I enjoy this channel due to your dev-centric content. I enjoy the other channels but their ‘creatives’ aspect always leaves me feeling like I want more. Keep these up I very much appreciate them.
@@RickertBrandsen This is a laptop, is not fair to test with chargers. OK for desktop machines, but laptop is made to work on battery, and there M kills everything.
@@Stratijev do u guys really live in a place where no chargers are around thus being said i have been sticking to the charger with my M1 pro most of the time it can stay on with my work flow for 2 hours max only
3:14 Just an FYI, when you're running docker on windows you're required to have WSL installed before you can even use the program. The reason why is because docker itself is a linux program. When you run docker on both windows and macbook, you're running a small vm of linux that allows you to be able to run docker as a program, that give or take except for the past year or so is normally worse in macbook's because there is more involved with translating the architecture for the new M1/M2 chips. You won't notice these on these tests because for simple Dockerfile's that are just use a base image of a linux distro they've likely been optimized to run on the arm architecture, but for a company's setup, they may/may not be set up to be able to run arm architecture on their computers.WSL runs near native performance for linux operations, but is still running under the hood a VM of a linux distro to run docker in the subsystem for you. Basically, it doesn't matter if you run it through windows or through your WSL distro, because the operation of docker runs through WSL. If it was running threw windows I assure you it would be minutes slower, if it even runs at all as windows distributions in general for docker have more bloated software, and do not run natively on windows platforms. This is also to say that this test is a bit meaningless as it really depends on your company's setup for docker. I know this as the last place I worked had significant issues with docker for over a year when the M1 intiially came out and people upgraded, many of the distros were not supporting arm architecture at the time and our setup was not built to be able to support that architecture. If you're just running some pet projects with minimal docker setup that you're likely controllling, then the Air is likely better for those operations, but if your grabbing this for an older codebase, I would highly encourage actually looking through your companys dockerfile's and if possible testing if it can even build the docker image, as you may not be able to. Things are much better now, as Macbook's recently came up with the rosetta translation solution that was added into docker a while ago that dramatically improves the translation layer between linux distros if you're company does not have arm support, but there still may be things that may not be supported yet for M1 laptops.
Docker can run windows containers on windows natively, without a VM behind the scene. It would be interesting to measure performance of this build in windows container.
Edit for the video: XPS is running 14 Physical Cores (20 Logical threads). vs MB running 8 Physical Cores (Arm doesn't do Hyperthreading, so 8 threads.)
*Random comment said they last 2 hours & need a charger 24/7 on their computer* it’s NOT about “not having a charger nearby”, It IS about, MacBooks being LAPTOPS & PORTABLE PERSONAL COMPUTERS and they LAST HOURS. I have a M1 TOUCHBAR 13” and MY music production workflow lasts 8+ HOURS. I WATCH 1 MOVIE it goes down 10% MAX?? That’s pretty good. For MY M1
I think a more fair comparison would be to the M2 MacBook Pro 14”. The Air is a compromise machine between portability and power. It is beautiful, but it isn’t competing for the pro market.
@@AZisk On a completely separate note. I just moved from two studio displays to a single Pro Display XDR, it's so nice having the extra space for the simulator on a single 32" screen. I wish I'd gotten it back when it first came out as it sounds like there'll be a new one in the next year or two, but still very happy. Also means I'm able to use the MacBook Air without DisplayLink work-arounds as I just use a single display now
this is the grim reality with this x86 generation. You have to spend a premium 2.6k to have roughly the same performance of the base macbook, while having half the battery time.
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 I just checked, in my country (Italy) the least possible price is 1.7k with 13700H, arc gpu, 16gb, windows home, 512MB and FHD monitor. Macbook Air 15 16GB is 2.1k. It's true, price/performance is really closer, if you don't need battery time or the fancy lcd.
It's pretty easy to find the XPS on sale for under $1500 but sales on all but the lowest end macs are almost non existent. Looking at Dells website now, I can see an XPS 15 with the i7 13700H for $1599 now as a BTS sale.
I would love to see test that's actually relevant to a lot of developers - quick builds every 30s or so for a prolonged period of time. Like you are working building website in typescript for example. I love my Air but it's performance over hours of short bursts was a concern for me.
For me, working on Jira at Atlassian, my Dell Precision (running Linux and the older 12th gen i7) compiles the project about 15% faster than my M1 MacBook pro and does fs IO bound operations 10 times faster (no exaggeration). yarn install takes 1/10th the time (when cached), git clean -xdf takes 1/10th the time. Rust compilation is about 50% faster on the Dell and Docker workflows are not even in the same universe (Dell running Linux decimates the MBP for Docker). Live recompiling for web (TypeScript) is noticeably faster on the Dell and interestingly, the MBP gets the same battery life as the Dell when doing live reloading work (~4-5 hours). My biggest criticism is the Dell is so horrible to use when compared to my MBP. The Dell trackpad is fatiguing to use (like doesn't Dell have a QA team, have they ever tried a MacBook?), where I can (and do) use the MBP in a cafe all day with the trackpad alone. The speakers are so garbage I wondered if they were broken. The Dell keyboard is fine and the screen is great. Despite the superior performance of the Dell, I reverted back to the slower MBP because I work from cafes/libraries a lot and portability/usability is more important to me. Though I did need to buy an ROG Ally to play games when I travel because the MBP can't play games. I also really really miss Linux. MacOS is nice, but Linux is really great for development.
I can’t imagine using any laptop for my development work but if I did it wouldn’t be a MacBook Air. A 16” MBP seems much more suitable. For side projects I use a 27” iMac but for work my employer provided a desktop tower with a water cooled 32 core Thread Ripper and two 49” ultra wide monitors.
you don't need a 1k machien to build a website you can do it with your chrome book same shit also to state that docker, sorting running varouise languages is not relevant to developers what are you writing HTML css js only all day?
@@andrewgrant788 I'm a react native develop, sometimes I do python backend too. I use macbook air 15" base model when I'm mobile and Mac Mini M2 Pro for my home setup. To be honest mba 15" is plenty powerful for my use case. I never managed to make it throttle throughout my dev session, and the battery is just awesome
Do not like my XPS 15. Runs hot with just a few MS apps opened. Might be they way my work set the battery and graphics settings. But every time I use it the keyboard gets uncomfortable to type. All my M chip macbooks stay fresh.
The operating system is a big reason. I don’t have the same experience under Linux as under windows. I can confirm that under windows the laptop is always hot in the touch …
@@ko-Daegu What are you doing with it? Is it the 14pro? or 13pro? or 16pro? I know I have the 13 m1 air and cool to the touch. I also have the 16pro and still cool to the touch even when I doc it with 3 monitors. XPS 15 just laptop alone hot and with 3 monitors hooked up its burning up.
I would not buy the XPS 15 for many reasons, but the XPS 17 would be tempting. The MB Pro 16 M2 Max is the obvious rival to the XPS 15 or 17 and is likely to be closer in performance than the Air due to more CPU and GPU cores. Out of these two, easy win for the Air, but probably the XPS 17 if that was the choice!
The XPS 17 is probably better overall, but the main difference in most of my use case scenarios is just the larger size. I don't need 100% of my i7/RTX card's performance beyond shorter bursts, which won't differ much in performance between the 15 and 17. I would never game on an XPS as it's always been a terrible experience throughout the history of this line (extremely slow displays) and I don't do a ton of heavy rendering or anything. And the 17 inch size does severely affect the portability.
Thanks for the video! I really disliked my XPS 15. It was loud, slow, had a horrible trackpad, and I needed a suitcase to carry the power brick. This was two years ago so they probably have fixed the issues. It really soured me on the XPS line. I've been leaning toward the LG Gram or Lenovo X1 series.
Hello Alex, Always enjoy your comparisons and reviews. It would be fair to make the same comparison with xps on Linux - docker is not(like never) native on Windows and Linux is closer to Macosx. Yes, Linux will require a lot of tinkering, drivers/fw are never as mature as on Windows, but it is where the fun begins.
This isn't true. WSL and mac are both running through a small VM that is allowing these OS to even be able to execute docker commands, since docker was not built natively for either windows or macs. The M1s may even be at a disadvantage in some cases as Arm architecture is not the standard binary that most images use, as they are mostly built for x86 architecture. There was an update a few months back to add in the rosetta translator that apple recently announced as a native feature into docker to allow it to have nearly the same performance as WSL for x86 images, but it's still inherently at a disadvantage in comparison to the XPS here.
Well, it's a bit unfair to compare a consumer device against the highest end device of Dell. I was surprised by the test with Docker, the C++ test shows practically same performance and where multicore matter, well, as you said, one has 20 cores, the other 8. I would say pretty good results for the MBA for 1000$ less, less weight, no fan, better keyboard, better touchpad, better sound and no reason to deal with Dell support 🙂 ah yes ... how performs the XPS running with the battery? 😛
Price difference is due to the specs he has. In Europe, XPS15 with 16GB RAM, 512GB SSD and RTX4050 costs less than my MBA15 with 16GB RAM and 512GB SSD. Having said that, Dell needs to change the design of the XPS15 and make it similar to the XPS13 Plus, with better keyboard, haptic touchpad, better cooling and better design.
I agree with all the point you said Dell needs to change: except the keyboard. For the target of this computer you cannot, in my opinion, lose the physical function keys. Anyway, price or not, the target of these two computer is different: the MBA is anyway supposed to target consumers, the other should be instead the highest end machine, so, I expect it’s faster than a machine that doesn’t neither have an active cooling system
@@gtangari Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
why did you pit a base level m2 processor with a H series intel core processor??? Please do a test comparing the vanilla m2 with either a U or P series processor 🙏
can you compare maybe the 15-inch MacBook air and laptops with 1360P/ 7735U Processors? or even gaming laptops with i5 13500H/7540HS.... because that is what most people would consider for 1K - 1.5K of the MacBook. A thin and light like the Asus S15 would also be a nice option, and I think you would get a result like " for 2/3rd the price, it offers better performance, while lacking a bit in build quality, and battery."
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
Most battery tests of laptops tend to focus on playing a video until it runs out. Please perform a "dev on a plane" battery test, e.g. continuously run a dotnet compile on a big project. Probably a worst case scenario, since you are not constantly compiling but will tell me how long I could be coding on a plane.
I dont code on any place i found, i sit for a while and usually plug it in, why drain the battery, I use a small 67w usb c charger, is almost iPhone size and it is enought
I wonder if 15-inch MBA M2 users can tell me whether or not you palm rest get remotely wam for normal usage like multimedia consumption, research, office work and light photo editing.
Hello Alex, love the channel. Regarding this video, I think we would benefit from benchmarks of real world projects and not just these quick algorithm runs.
That's i7 and the most powerfull mobile H chip so it's kinda not fair to compare it to air, it should be compared to 'U' or 'P' varients H should be compared to M2 pro chips honestly
It has 6 power cores that are hyper threaded and 8 efficiency cores, while m2 air has 4 power and 4 efficiency cores. M2 pro 12 core has 8 power and 4 efficiency cores and would obliterate the 13700h
Hi Alex, after having a MacBook M2 AIR 15' base variant (8/256), I am confused whether I could have bought 512 gb. Could you please help me if purchased base model is good for belo tasks in terms of MEMORY (256 GB): 1. Casually using docker 2. Terraform 3. Learning other DevOps activities like Jenkins 4. No video editing Above is the usage please help!
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
Thanks for showing that MBA is the right choice for developer. Even it is 1% slower at certain tasks than the ugly low quality heavy hot energy inefficient and EXPENSIVE Dell XPS.
I would like to see compilation of GCC in one test. Another test where 3 VM's are run at the same time. Next do some synthesis of an fpga build. And finally a soldworks test and maybe a 3d simulation. I feel like these tests would let a engineer get a better feel between the 2 laptops.
Guys pls help I have 3 more years of high school and 1300 bucks. I currently am using a base model 2017 MacBook Air. Should I get the 2020 M1 MacBook Air or keep mine?
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 seems like you're so desperate to prove your point, you've missed mine. The M2 Pro MacBook Pro, at the same price-point as the PC he tested, has much higher CPU performance which would have made for a better comparison. Also, cheap, performant hardware won't fix Windows - ever.
@@daveh6356 at 2600 yes, but at 1500, 1200 dollar laptops, dell is better, Apple has bad cheap laptops, they are good but lack performamce. Also, where they lack even at 2600 is gpu, the 4050 beats at the 1500 mark and at 2600 the 4070 wins, specially gaming
Main problem with powerful Windows laptop is that they're way more expensive than Apple offering. As someone who really want CUDA and OptiX and Intel multi-threaded performance, the price is simply way out of reach. (Man can't believe I said this on Intel products), And on Apple specifically, Metal Acceleration does help but it is hundred miles behind Nvidia in terms of performance even on something like M1/M2 Ultra Mac Studio.
@@farid.2972 I'm fully aware of that but as someone who are using M2 MacBook Air as a daily driver there's still lot more than just "performance on battery". I can't deny when I'm doing video editing all the available processing power is there even on battery but when I'm trying to access program that can make use of CUDA i.e PyTorch, Macbook is way too slow for that task even when I set the program to run on MPS, As for Windows, yes you'll hardly get any performance on Nvidia GPU on battery but at least you know that it is there as soon as you plug it to a power brick, when you want it.
Never in my life will buy another Dell, worst experience in my life. My XPS 15 9500 have lot of problems 1. Some hardware problem that in the lapse of 2 days the SSD disconnects and shows an error screen. 2. Battery inflated and trackpad stopped responding(now without battery) 3. Some screen frezzes when external monitor used(the computer keep running, can hear sounds of interactions with mouse ) 4. Audio hardware is trash, in windows the volume of teams is very very low even at full volume but the other sounds are incredible loud. On Linux the mic is trash, it’s like talking from inside a blender.
i see you are running the windows side of test on ubnutu which has some translation overhead. i think running Linux on dual boot will have much more performance.
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
I spent last week with M2 Pro 14 inch and... going back to T14 with Ryzen. Yes it's way slower, worse build quality etc, but man. MacOS makes the MacBook unusable for me. I made sure I will give myself a week to get used to it, but no. If you use it on a battery, on a balcony it's great. But as soon as you plug in a keyboard and mouse, external screen, the OS starts to show its colors. Why THE FFFFF do I have to click on a window to activate it first? That's hundreds of clicks per day and makes two screens useless. Why do I have to install 5 apps to make multitasking somewhat ok? The dock is a joke - no way to get two icons for two chrome profiles? no notifications on the dock? and why the traffic light buttons are in a slighlty different spot in each app? Oh, did I mention that you need external app to have the mouse scrolling the correct way and keep touchpad natural scroll at the same time? It's really sad that such great hardware is hindered by the software.
If you are gonna do that then use AlDente to stop charging over a certain limit. I would suggest 50-60%. There is no question that charging constantly at 100% is going to reduce the life of the battery as Apple does not use charge-through technology. Charging to 80% every time rather than 100% is likely to add 1000 full cycles to your battery life according to research.
@@maskedroyale Try and keep your battery between 20-85% if you use it on battery. But if it is only on mains then stop charging around 50%. AlDente is a macOS app that does this. My 2.5-year-old MB Pro 13 M1 still has 99% battery capacity!
@@maskedroyale I never use my MacBook on power. Every evening I top it up to 85% or around that mark. I use it for about 6 to 7 hours a day during the week. And rarely does it go below 30%. That way I keep the battery at its optimum zone preserving the max battery cycles. Obviously, people have different uses so they may not be able to manage the battery like I do, but as the MacBook Air is an ultrabook, most people could use it like that.
Any reductionist type of test like this is misleading. Faster does not always translate to better user experience in real world scenarios. Without considering the other unavoidable aspects of using each machine. I have owned two Dell XPS laptops a 13" and 15" both upgraded (not base models). Both had numerous issues including, charging, battery life, trackpad, audio jack, rebooting, and overheating. These are not issues that are due to me getting lemons, as the forums are full of similar complaints from other users that have been going on for many years. Dell is well aware of these issues, but for whatever reason, will not resolve them. I won't list the issues and hassles that arose as a consequence of using Microsoft Windows. It is my opinion that there is inevitably going to be detrimental consequences, when you have one company creating the hardware and another creating the operating system. Naturally, each company is not going to be fully transparent or fully cooperative with the other company even if they recognize it is a symbiotic relationship. While I am sure there is some segregation and separation with Apple's development processes, it would not be as pronounced as with the prior scenario.
Depends, because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
XPS 15 cost is as much as 2 MacBook pro plus XPS only works in cold countries like US and Europe. XPS processor will get damaged at 40 to 50 degree summer room temperature.
Stop spreading misinformation. I'm from Brazil (a very hot country), own an XPS 15 with an i9 and the CPU cores are always at 29-38° when doing light tasks and 50-80° when gaming or something more demanding.
I would recommend a Blade 16 or Legion Pro over the XPS. Similar prices (Razer a bit more). The display on the XPS are lagging behind, not efficient cooling for CPU + dedicated GPU.
@@Luan-RT I live in Germany and I had an XPS15 with i9 and no man, I live in a "cold" country, but the notebook was getting super hot immediately (and yes, throttling if you don't set the appropriate fan configuration which btw means: *noises*). My notebook also had the well known (but not recognized by Dell) problem with the hinges and at the trackpad. Dell: no more, thanks :)
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@divitvaghani158 yes I have, still every windows laptop from different companies has its own few good things about it, though no Laptops unite all those features in a single device like the MacBooks
Keep in mind intel is still on 10nm too; impressive how performant it is compared to 5nm arm SoC. Im really interested to see how power efficient and powerful the x86 chips from intel are going to be when they get to 5nm.
The XPS hides muscle under that beautiful finish, but I think that in real use it would end up having heat issues, I still love it. Have you tried the Asus Zenbook Pro 16X OLED w/ i9 13th gen & RTX 4070? It's the same price as the XPS in this video, and its reviews are excellent.
Why some think that developer wants only the fastest CPU on Earth? Developers are just the common people like the others. They want a quality tool for a job, a tool that they can rely on.
@@AZisk unfortunately I need(ed) to deal a lots with Windows notebooks and rarely Dell have been the best (the only one ever satisfying me was the XPS13 3 years ago). In my opinion, the best Windows laptops are still the ThinkPads: the X1 (Extreme) in particular.
Apple's Soccer Mom / Starbucks Squatter $1600.00 laptop (8 core/16GB) holds its own against a Dell XPS $2600.00 (20 core/32 GB RAM) x86-based Developer-class laptop. Yawn....what else is new ??
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 Nope ...The 1200 and 1500 Dollar version of this Dell would get BLOWN AWAY by the MBA. That's why he had to use the "upgraded" Unit to make it even close to a fair fight. Not worth anywhere near the $2600 asking price. BTW, I would rather use a base M1 MBA than ANY flagship x86-based laptop...anyday.
@@goobfilmcast4239 this test dont use the gpu, only cpu, the result will be the same, because the cpu is the same, the i7-13700h, also from 16 to 32gb ram in this case doesnt make any change
![Eminem - Fuel (feat. JID) [Official Lyric Video]](http://i.ytimg.com/vi/t5H_CewqpKA/mqdefault.jpg)
I enjoy this channel due to your dev-centric content. I enjoy the other channels but their ‘creatives’ aspect always leaves me feeling like I want more. Keep these up I very much appreciate them.
It would be interesting to see the results with no charger plugged in
that was already done i think and macs are winning in this.
@@RickertBrandsen This is a laptop, is not fair to test with chargers. OK for desktop machines, but laptop is made to work on battery, and there M kills everything.
@@Stratijev didn't I say it, lol?
@@Stratijev do u guys really live in a place where no chargers are around thus being said i have been sticking to the charger with my M1 pro most of the time
it can stay on with my work flow for 2 hours max only
@@ko-Daegu
Nah not possible am using mba m1 and I get more than 8 hours with full potential
3:14 Just an FYI, when you're running docker on windows you're required to have WSL installed before you can even use the program. The reason why is because docker itself is a linux program. When you run docker on both windows and macbook, you're running a small vm of linux that allows you to be able to run docker as a program, that give or take except for the past year or so is normally worse in macbook's because there is more involved with translating the architecture for the new M1/M2 chips.
You won't notice these on these tests because for simple Dockerfile's that are just use a base image of a linux distro they've likely been optimized to run on the arm architecture, but for a company's setup, they may/may not be set up to be able to run arm architecture on their computers.WSL runs near native performance for linux operations, but is still running under the hood a VM of a linux distro to run docker in the subsystem for you.
Basically, it doesn't matter if you run it through windows or through your WSL distro, because the operation of docker runs through WSL. If it was running threw windows I assure you it would be minutes slower, if it even runs at all as windows distributions in general for docker have more bloated software, and do not run natively on windows platforms.
This is also to say that this test is a bit meaningless as it really depends on your company's setup for docker. I know this as the last place I worked had significant issues with docker for over a year when the M1 intiially came out and people upgraded, many of the distros were not supporting arm architecture at the time and our setup was not built to be able to support that architecture. If you're just running some pet projects with minimal docker setup that you're likely controllling, then the Air is likely better for those operations, but if your grabbing this for an older codebase, I would highly encourage actually looking through your companys dockerfile's and if possible testing if it can even build the docker image, as you may not be able to.
Things are much better now, as Macbook's recently came up with the rosetta translation solution that was added into docker a while ago that dramatically improves the translation layer between linux distros if you're company does not have arm support, but there still may be things that may not be supported yet for M1 laptops.
Docker can run windows containers on windows natively, without a VM behind the scene. It would be interesting to measure performance of this build in windows container.
Windows is a pain in the a** in terms of resource usage : even idling the laptop runs hot and fans are running …
My lenovo with ryzen isn't hot. Fan is stopped
Edit for the video: XPS is running 14 Physical Cores (20 Logical threads). vs MB running 8 Physical Cores (Arm doesn't do Hyperthreading, so 8 threads.)
*Random comment said they last 2 hours & need a charger 24/7 on their computer*
it’s NOT about “not having a charger nearby”,
It IS about, MacBooks being LAPTOPS & PORTABLE PERSONAL COMPUTERS and they LAST HOURS. I have a M1 TOUCHBAR 13” and MY music production workflow lasts 8+ HOURS.
I WATCH 1 MOVIE it goes down 10% MAX?? That’s pretty good.
For MY M1
I think a more fair comparison would be to the M2 MacBook Pro 14”. The Air is a compromise machine between portability and power. It is beautiful, but it isn’t competing for the pro market.
The air can still beat the XPS 15 on battery
The 15-inch Air compares to the Lenovo ThinkPad T16 series or LG Gram 16-inch.
M2 Air >> when it comes to performance per $.
M1 Pro >> in performance at a still lower price.
Thanks for the video - it'd be nice to also see some tests without the laptop plugged in.
ooh that wouldn’t be good, no no no. a windows laptop must always be plugged in 😆
@@AZisk On a completely separate note. I just moved from two studio displays to a single Pro Display XDR, it's so nice having the extra space for the simulator on a single 32" screen. I wish I'd gotten it back when it first came out as it sounds like there'll be a new one in the next year or two, but still very happy. Also means I'm able to use the MacBook Air without DisplayLink work-arounds as I just use a single display now
@@rhoderzau that sounds like a really nice screen. I'm still in the sub-1k range for monitors, ruining my eyes over here :)
@@AZisk haha I overrode the 500 nit limit so I'm running my eyes in a different way.
Thanks for the video, it would be really cool to see more comparisons of the same tests when the laptops are running on battery!
this is the grim reality with this x86 generation. You have to spend a premium 2.6k to have roughly the same performance of the base macbook, while having half the battery time.
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 I just checked, in my country (Italy) the least possible price is 1.7k with 13700H, arc gpu, 16gb, windows home, 512MB and FHD monitor.
Macbook Air 15 16GB is 2.1k.
It's true, price/performance is really closer, if you don't need battery time or the fancy lcd.
It's pretty easy to find the XPS on sale for under $1500 but sales on all but the lowest end macs are almost non existent. Looking at Dells website now, I can see an XPS 15 with the i7 13700H for $1599 now as a BTS sale.
with terrible screen resolution
I would love to see test that's actually relevant to a lot of developers - quick builds every 30s or so for a prolonged period of time. Like you are working building website in typescript for example. I love my Air but it's performance over hours of short bursts was a concern for me.
For me, working on Jira at Atlassian, my Dell Precision (running Linux and the older 12th gen i7) compiles the project about 15% faster than my M1 MacBook pro and does fs IO bound operations 10 times faster (no exaggeration). yarn install takes 1/10th the time (when cached), git clean -xdf takes 1/10th the time. Rust compilation is about 50% faster on the Dell and Docker workflows are not even in the same universe (Dell running Linux decimates the MBP for Docker).
Live recompiling for web (TypeScript) is noticeably faster on the Dell and interestingly, the MBP gets the same battery life as the Dell when doing live reloading work (~4-5 hours).
My biggest criticism is the Dell is so horrible to use when compared to my MBP. The Dell trackpad is fatiguing to use (like doesn't Dell have a QA team, have they ever tried a MacBook?), where I can (and do) use the MBP in a cafe all day with the trackpad alone. The speakers are so garbage I wondered if they were broken. The Dell keyboard is fine and the screen is great.
Despite the superior performance of the Dell, I reverted back to the slower MBP because I work from cafes/libraries a lot and portability/usability is more important to me. Though I did need to buy an ROG Ally to play games when I travel because the MBP can't play games.
I also really really miss Linux. MacOS is nice, but Linux is really great for development.
I can’t imagine using any laptop for my development work but if I did it wouldn’t be a MacBook Air. A 16” MBP seems much more suitable. For side projects I use a 27” iMac but for work my employer provided a desktop tower with a water cooled 32 core Thread Ripper and two 49” ultra wide monitors.
you don't need a 1k machien to build a website you can do it with your chrome book same shit
also to state that docker, sorting running varouise languages is not relevant to developers what are you writing HTML css js only all day?
@@ko-Daegu some people do write just js/css/html. A lot of devs don't use docker
@@andrewgrant788 I'm a react native develop, sometimes I do python backend too. I use macbook air 15" base model when I'm mobile and Mac Mini M2 Pro for my home setup. To be honest mba 15" is plenty powerful for my use case. I never managed to make it throttle throughout my dev session, and the battery is just awesome
Do not like my XPS 15. Runs hot with just a few MS apps opened. Might be they way my work set the battery and graphics settings. But every time I use it the keyboard gets uncomfortable to type. All my M chip macbooks stay fresh.
The operating system is a big reason. I don’t have the same experience under Linux as under windows. I can confirm that under windows the laptop is always hot in the touch …
my macbook m1 pro get's hot nost of the day too
i feel i was sold to the hype
@@ko-Daegu What are you doing with it? Is it the 14pro? or 13pro? or 16pro? I know I have the 13 m1 air and cool to the touch. I also have the 16pro and still cool to the touch even when I doc it with 3 monitors. XPS 15 just laptop alone hot and with 3 monitors hooked up its burning up.
I would not buy the XPS 15 for many reasons, but the XPS 17 would be tempting. The MB Pro 16 M2 Max is the obvious rival to the XPS 15 or 17 and is likely to be closer in performance than the Air due to more CPU and GPU cores. Out of these two, easy win for the Air, but probably the XPS 17 if that was the choice!
The XPS 17 is probably better overall, but the main difference in most of my use case scenarios is just the larger size. I don't need 100% of my i7/RTX card's performance beyond shorter bursts, which won't differ much in performance between the 15 and 17. I would never game on an XPS as it's always been a terrible experience throughout the history of this line (extremely slow displays) and I don't do a ton of heavy rendering or anything. And the 17 inch size does severely affect the portability.
I’d say the main difference is Performance Per Watt, the XPS will run much hotter and louder, and drain your battery at least 2x faster.
I will wait for next year, with intel meteor lake cpu that is focused on efficiency and hope for a refresh or redesign with vapour chamber cooling
Thanks for the video! I really disliked my XPS 15. It was loud, slow, had a horrible trackpad, and I needed a suitcase to carry the power brick. This was two years ago so they probably have fixed the issues. It really soured me on the XPS line. I've been leaning toward the LG Gram or Lenovo X1 series.
Hello Alex,
Always enjoy your comparisons and reviews.
It would be fair to make the same comparison with xps on Linux - docker is not(like never) native on Windows and Linux is closer to Macosx.
Yes, Linux will require a lot of tinkering, drivers/fw are never as mature as on Windows, but it is where the fun begins.
This isn't true. WSL and mac are both running through a small VM that is allowing these OS to even be able to execute docker commands, since docker was not built natively for either windows or macs. The M1s may even be at a disadvantage in some cases as Arm architecture is not the standard binary that most images use, as they are mostly built for x86 architecture. There was an update a few months back to add in the rosetta translator that apple recently announced as a native feature into docker to allow it to have nearly the same performance as WSL for x86 images, but it's still inherently at a disadvantage in comparison to the XPS here.
Why you have to care about a $3 trillion company software ecosystem?
mark some test using orbstack instead of docker compose , amazing increase
Well, it's a bit unfair to compare a consumer device against the highest end device of Dell. I was surprised by the test with Docker, the C++ test shows practically same performance and where multicore matter, well, as you said, one has 20 cores, the other 8. I would say pretty good results for the MBA for 1000$ less, less weight, no fan, better keyboard, better touchpad, better sound and no reason to deal with Dell support 🙂
ah yes ... how performs the XPS running with the battery? 😛
Price difference is due to the specs he has. In Europe, XPS15 with 16GB RAM, 512GB SSD and RTX4050 costs less than my MBA15 with 16GB RAM and 512GB SSD.
Having said that, Dell needs to change the design of the XPS15 and make it similar to the XPS13 Plus, with better keyboard, haptic touchpad, better cooling and better design.
I agree with all the point you said Dell needs to change: except the keyboard. For the target of this computer you cannot, in my opinion, lose the physical function keys.
Anyway, price or not, the target of these two computer is different: the MBA is anyway supposed to target consumers, the other should be instead the highest end machine, so, I expect it’s faster than a machine that doesn’t neither have an active cooling system
@@gtangari Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
its Schwarzenegger for me. got all advanced now, thats progress in my book....
why did you pit a base level m2 processor with a H series intel core processor??? Please do a test comparing the vanilla m2 with either a U or P series processor 🙏
I believe that the pyrometer should be thrown into the trash and buy a normal thermal imager. Also need to check coil whine noise on both laptops!!!
Did you connected the dell xps to charger while doing tests 🙄?
Yeah, the charger icon shows its not connected🤔
The MacBook Air gets the the price-to-performance award.
can you compare maybe the 15-inch MacBook air and laptops with 1360P/ 7735U Processors? or even gaming laptops with i5 13500H/7540HS.... because that is what most people would consider for 1K - 1.5K of the MacBook. A thin and light like the Asus S15 would also be a nice option, and I think you would get a result like " for 2/3rd the price, it offers better performance, while lacking a bit in build quality, and battery."
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
Most battery tests of laptops tend to focus on playing a video until it runs out.
Please perform a "dev on a plane" battery test, e.g. continuously run a dotnet compile on a big project. Probably a worst case scenario, since you are not constantly compiling but will tell me how long I could be coding on a plane.
I dont code on any place i found, i sit for a while and usually plug it in, why drain the battery, I use a small 67w usb c charger, is almost iPhone size and it is enought
Did you also do this test against the Macbook pro M2 Max?
I wonder if 15-inch MBA M2 users can tell me whether or not you palm rest get remotely wam for normal usage like multimedia consumption, research, office work and light photo editing.
Nice vidéo simple and efficient it would be great to have a comparison on php with zbench for instance
Hello Alex, love the channel. Regarding this video, I think we would benefit from benchmarks of real world projects and not just these quick algorithm runs.
Battery operation? Battery life? I knight have missed it: did you use clang on the air?
That's i7 and the most powerfull mobile H chip so it's kinda not fair to compare it to air, it should be compared to 'U' or 'P' varients
H should be compared to M2 pro chips honestly
It has 6 power cores that are hyper threaded and 8 efficiency cores, while m2 air has 4 power and 4 efficiency cores.
M2 pro 12 core has 8 power and 4 efficiency cores and would obliterate the 13700h
Hi Alex, after having a MacBook M2 AIR 15' base variant (8/256), I am confused whether I could have bought 512 gb. Could you please help me if purchased base model is good for belo tasks in terms of MEMORY (256 GB):
1. Casually using docker
2. Terraform
3. Learning other DevOps activities like Jenkins
4. No video editing
Above is the usage please help!
8gb in 2023 it's not enough !
isn't the XPS 15 expensive compared to Macbook Air? Wouldn't a fair comparison be between M1 Pro/M2 Pro chip macs?
In Europe, MBA15 with 16GB RAM and 512GB storage is more expensive than the XPS15 i7 16GB RAM, 512GB SSD and RTX4050
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
Thanks for showing that MBA is the right choice for developer. Even it is 1% slower at certain tasks than the ugly low quality heavy hot energy inefficient and EXPENSIVE Dell XPS.
Do this with a Zephyrus G14 2023 vs a 2022!
I think you should do benchmarks on ubuntu vs max, as windows is lot slower, and there is dell developer editon on xps
What about fan noise for XPS 15?
If I will install a bare metal linux like ubuntu, should I still go for Mac?
I would like to see compilation of GCC in one test. Another test where 3 VM's are run at the same time. Next do some synthesis of an fpga build. And finally a soldworks test and maybe a 3d simulation. I feel like these tests would let a engineer get a better feel between the 2 laptops.
Guys pls help I have 3 more years of high school and 1300 bucks. I currently am using a base model 2017 MacBook Air. Should I get the 2020 M1 MacBook Air or keep mine?
Thanks!
🤩 thanks !
I love my Air but seriously a 20-core $2600 machine against an 8-core $1600 machine? Try the 12-core M2 Pro.
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 seems like you're so desperate to prove your point, you've missed mine. The M2 Pro MacBook Pro, at the same price-point as the PC he tested, has much higher CPU performance which would have made for a better comparison.
Also, cheap, performant hardware won't fix Windows - ever.
@@daveh6356 at 2600 yes, but at 1500, 1200 dollar laptops, dell is better, Apple has bad cheap laptops, they are good but lack performamce. Also, where they lack even at 2600 is gpu, the 4050 beats at the 1500 mark and at 2600 the 4070 wins, specially gaming
MacBook Air 15" is still plenty fast enough for developers and stays cool etc.
Hey Max, still waiting on the android development for the M2 Max & pro :-)
just got xps 15 today
congrats 🎊
You would also need ice packs to cool that heater 😀
Congrats! Ignore the applers....
can other laptops with the same hardware will perform the same?
did you used wsl2 on windows as docker engine or hyper-v ?
Hello, is it possible to enable low power mode on m2 pro models and m2 max models or it just for basic m2 chips models?
How do you have those seconds timers for each line on the terminal?
Main problem with powerful Windows laptop is that they're way more expensive than Apple offering. As someone who really want CUDA and OptiX and Intel multi-threaded performance, the price is simply way out of reach. (Man can't believe I said this on Intel products),
And on Apple specifically, Metal Acceleration does help but it is hundred miles behind Nvidia in terms of performance even on something like M1/M2 Ultra Mac Studio.
One of the biggest downsides of x86 machines is that they have not the same performance on battery compared to macs
@@farid.2972 I'm fully aware of that but as someone who are using M2 MacBook Air as a daily driver there's still lot more than just "performance on battery".
I can't deny when I'm doing video editing all the available processing power is there even on battery but when I'm trying to access program that can make use of CUDA i.e PyTorch, Macbook is way too slow for that task even when I set the program to run on MPS,
As for Windows, yes you'll hardly get any performance on Nvidia GPU on battery but at least you know that it is there as soon as you plug it to a power brick, when you want it.
Maybe for you.. Linux is the way to go
@@cristianarean647 I’m not talking about software. I’m talking about hardware acceleration
Never in my life will buy another Dell, worst experience in my life. My XPS 15 9500 have lot of problems
1. Some hardware problem that in the lapse of 2 days the SSD disconnects and shows an error screen.
2. Battery inflated and trackpad stopped responding(now without battery)
3. Some screen frezzes when external monitor used(the computer keep running, can hear sounds of interactions with mouse )
4. Audio hardware is trash, in windows the volume of teams is very very low even at full volume but the other sounds are incredible loud. On Linux the mic is trash, it’s like talking from inside a blender.
What is website you are use to benchmark python?
Can you provide url
i see you are running the windows side of test on ubnutu which has some translation overhead. i think running Linux on dual boot will have much more performance.
he literally showed in the video that there is no performance difference between the wsl - ubuntu terminal and the windows terminal
It would be reasonable to match the xps with the same price class e.g: mbp 16 inch m2 + unplugged for the xps
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
I spent last week with M2 Pro 14 inch and... going back to T14 with Ryzen. Yes it's way slower, worse build quality etc, but man. MacOS makes the MacBook unusable for me. I made sure I will give myself a week to get used to it, but no. If you use it on a battery, on a balcony it's great. But as soon as you plug in a keyboard and mouse, external screen, the OS starts to show its colors.
Why THE FFFFF do I have to click on a window to activate it first? That's hundreds of clicks per day and makes two screens useless. Why do I have to install 5 apps to make multitasking somewhat ok? The dock is a joke - no way to get two icons for two chrome profiles? no notifications on the dock? and why the traffic light buttons are in a slighlty different spot in each app? Oh, did I mention that you need external app to have the mouse scrolling the correct way and keep touchpad natural scroll at the same time?
It's really sad that such great hardware is hindered by the software.
Hey Alex 1 question, does it hamper the battery life of MacBook the new apple silicon MacBooks if we use it continuously plugged into the socket?
If you are gonna do that then use AlDente to stop charging over a certain limit. I would suggest 50-60%. There is no question that charging constantly at 100% is going to reduce the life of the battery as Apple does not use charge-through technology. Charging to 80% every time rather than 100% is likely to add 1000 full cycles to your battery life according to research.
@@andyH_England According to you what is going to be the best way to keep the battery life and health good?
@@maskedroyale Try and keep your battery between 20-85% if you use it on battery. But if it is only on mains then stop charging around 50%. AlDente is a macOS app that does this. My 2.5-year-old MB Pro 13 M1 still has 99% battery capacity!
@@andyH_England How do you charge your laptop? How do you use your laptop plugged in or on battery? Your routine.
@@maskedroyale
I never use my MacBook on power. Every evening I top it up to 85% or around that mark. I use it for about 6 to 7 hours a day during the week. And rarely does it go below 30%. That way I keep the battery at its optimum zone preserving the max battery cycles. Obviously, people have different uses so they may not be able to manage the battery like I do, but as the MacBook Air is an ultrabook, most people could use it like that.
Apples to oranges IMHO. A developer would never go for an Air as the main machine but for a MBP.
How dare you call my Dell an orange!
Hey ... don't bash orange. It's my national colour ... @@AZisk
Any reductionist type of test like this is misleading. Faster does not always translate to better user experience in real world scenarios. Without considering the other unavoidable aspects of using each machine. I have owned two Dell XPS laptops a 13" and 15" both upgraded (not base models). Both had numerous issues including, charging, battery life, trackpad, audio jack, rebooting, and overheating. These are not issues that are due to me getting lemons, as the forums are full of similar complaints from other users that have been going on for many years. Dell is well aware of these issues, but for whatever reason, will not resolve them. I won't list the issues and hassles that arose as a consequence of using Microsoft Windows. It is my opinion that there is inevitably going to be detrimental consequences, when you have one company creating the hardware and another creating the operating system. Naturally, each company is not going to be fully transparent or fully cooperative with the other company even if they recognize it is a symbiotic relationship. While I am sure there is some segregation and separation with Apple's development processes, it would not be as pronounced as with the prior scenario.
Which metrics are best for video editing? 📈 📸
how will it be different without ac?
The Air catches up fast when you factor in the time you'll spend struggling with Windows Updates, LOL
Would it be safe to say the XPS 15 results would be similar for any i7-13700 equipped Windows laptop?
If not better, because the xps runs the cpu at 60w, other laptops can go up to 100-120w
Oh, No
Can you test with ryzen 7040 series? Like 7840hs
Great, but I think that the MacBook Air M2 is better (value/performance) especially if you used MacBook Air M2 14” model which is cost less
Depends, because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
It isn't fair to compare two notebooks with different number of cores.
No, only thing to check is size, weight and price on a laptop
Xps get hot quickly
XPS 15 cost is as much as 2 MacBook pro plus XPS only works in cold countries like US and Europe. XPS processor will get damaged at 40 to 50 degree summer room temperature.
Stop spreading misinformation. I'm from Brazil (a very hot country), own an XPS 15 with an i9 and the CPU cores are always at 29-38° when doing light tasks and 50-80° when gaming or something more demanding.
I would recommend a Blade 16 or Legion Pro over the XPS. Similar prices (Razer a bit more). The display on the XPS are lagging behind, not efficient cooling for CPU + dedicated GPU.
@@Luan-RT I live in Germany and I had an XPS15 with i9 and no man, I live in a "cold" country, but the notebook was getting super hot immediately (and yes, throttling if you don't set the appropriate fan configuration which btw means: *noises*). My notebook also had the well known (but not recognized by Dell) problem with the hinges and at the trackpad. Dell: no more, thanks :)
@@Luan-RT I am talking about outside temperature of 40 to 50 degree and not core temperature
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
the air wins because the dell cant run xcode....
I would never use Windows Laptops, just Macs
Why?)
@@vasylzhydun8879 cause windows is literally sh*t, and the laptops suck, they’re slow
@@mr.morales0707😂😂 I guess you haven't use very high end machine
@@divitvaghani158 yes I have, still every windows laptop from different companies has its own few good things about it, though no Laptops unite all those features in a single device like the MacBooks
@@mr.morales0707slow ? Bro what windows laptop you using, omg typical apple fanboy behaviour
Keep in mind intel is still on 10nm too; impressive how performant it is compared to 5nm arm SoC. Im really interested to see how power efficient and powerful the x86 chips from intel are going to be when they get to 5nm.
So, what you meant is 5nm of banana is better than 10nm of apple 😂
Wait to meteor lake next year, around 10-20% more power and 20% more efficiency
@@broccoloodle what? Apple is the one at 5nm, intel is at 10nm, 10 is worse than 5 bro
We want mooooore
$1000 more for 10s faster performance? I choose macbook air.. because of the trackpad 😁
The XPS hides muscle under that beautiful finish, but I think that in real use it would end up having heat issues, I still love it.
Have you tried the Asus Zenbook Pro 16X OLED w/ i9 13th gen & RTX 4070? It's the same price as the XPS in this video, and its reviews are excellent.
In my experience WSL is generally faster than native Windows for most development workloads haha.
Unplug them and the Dell will suffer 😋
Why some think that developer wants only the fastest CPU on Earth? Developers are just the common people like the others. They want a quality tool for a job, a tool that they can rely on.
I love my xps15
it’s the best windows laptop imo
@@AZisk unfortunately I need(ed) to deal a lots with Windows notebooks and rarely Dell have been the best (the only one ever satisfying me was the XPS13 3 years ago). In my opinion, the best Windows laptops are still the ThinkPads: the X1 (Extreme) in particular.
I agree, I've tried many and it's a step above the others.
Apple's Soccer Mom / Starbucks Squatter $1600.00 laptop (8 core/16GB) holds its own against a Dell XPS $2600.00 (20 core/32 GB RAM) x86-based Developer-class laptop. Yawn....what else is new ??
XPS was running 14 Cores (20 logical threads)
@@ScottGrunwald well....that explains everything /s
Because he has the rtx 4070, with the rtx 4050 is 1500, and this test didnt use the gpu, you could also get the intel arc gpu that is way cheaper 1200 and get a cheaper faster cpu laptop than the mac
@@dracoxruiz2563 Nope ...The 1200 and 1500 Dollar version of this Dell would get BLOWN AWAY by the MBA. That's why he had to use the "upgraded" Unit to make it even close to a fair fight. Not worth anywhere near the $2600 asking price. BTW, I would rather use a base M1 MBA than ANY flagship x86-based laptop...anyday.
@@goobfilmcast4239 this test dont use the gpu, only cpu, the result will be the same, because the cpu is the same, the i7-13700h, also from 16 to 32gb ram in this case doesnt make any change
Can't build iOS apps on windows laptop
nope, sure can’t.
Third?
Second?
First?