Hi, Thanks for watching, glad you found it useful. Been wandering how the results differ for 14900k if you are able to please share your results would be great to know. All the best with the build, TK :).
Thanks. I just got my ALF and already have the contact frame installed, the Thermal Grizzly one. Was kind of nervous on mounting it with the all the pressure and such.
Hi, Thanks for watching. You are welcome glad you found it useful. To be honest I am in the same boat, don't normally attempt such mods but had to be attempted with the temps these CPU's attain. Just want to mention that I did actually have an issue when I first mounted it with the 13900k, and had to tighten the screws a little more, was scary that it crashed at first but has been stable ever since. Glad you managed to install it, which ALF did you get and which CPU do use, would be great to know what sort of temps you get if you have time. All the best, TK :).
@@TKsTechTalk I got the Arctic Liquid Freezer II 360, non RGB and the I7 13700K. I haven’t installed the new AIO yet, as it arrived missing hardware, few O-Rings. I’m currently using the 240 H100i Elite AIO from Corsair and the temps bottleneck when stress testing, so I had to undervolt the CPU, which helps for now.
@@haies09 Ok nice, that would be similar in TDP to 12900k but again the IHS seems to be improved on the 13th gen CPU's so I guess you should get decent results but maybe not as big a drop as 12th gen, and undervolting is def the way to go even with the AIO as you may have seen from my tests, without any real performance loss at all. All the best with the AIO upgrade :).
Thank you for this video. I currently have the Thermal Grizzly contact frame, but I feel like I have it too tight, because I’ve experienced some game crashes sporadically with a very light overclock and extremely good temps. However, I think my stock retention bracket screws are a tad too short for the thickness of this new motherboard and it doesn’t allow me to back the screws out to a lighter pressure without pulling the screws out of the backplate. I ordered the Thermalright LGA1700 contact frame yesterday and should have it by this afternoon hopefully so I can swap it out tonight and hopefully have more system/ram stability. My power limits have been removed and I have cores 1-5 at 5.9Ghz and cores 6-8 at 5.8Ghz. Looks to be around 1.43v at the worst(@ idle with windows 11 power plan set to high performance mode, which keeps the processor at 100% power at idle). My temps have never reached over about 75-80c in all the milder benchmarks(CPU-Z multi-core benchmark(71-75C) and stress test(80-81C) functions) I’ve done with an idle temp of around 26-30C depending on room temp/AC on. I have “SVID Behavior” set to “Trained Behavior” instead of Auto or Typical behavior. This helps keep the voltage and temps down compared to using “typical” or “worst case scenario”(intel’s model, which can pump up to 1.8v!!!)”, which will pump in way more voltage than needed and increase the temps. I set Load Line Calibration to level 4(Asus) to reduce Vdroop a little, which improves OC stability. I may bump this to level 6. Not sure yet. My system specs are as follows: -Intel 13900ks CPU -Asus ROG Maximus Z790 APEX (2-DIMM motherboard) -G.Skill 32GB 7800mhz XMP DDR5 kit -Corsair h170i elite lcd 420mm AIO(6 push/pull fans pulling in cool air) -Thermal Grizzly 12/13th gen contact frame(soon to be Thermalright version) -Samsung 990 Pro 2TB nvme SSD -Gigabyte Gaming OC 4090(stock 450watts) -Asus Thor 1600watt PSU -Corsair 1000D Super tower case(tons of room, but very heavy) -Windows 11 Home -Samsung Odyssey Neo G8 4k 240Hz monitor(I reduced this to 120Hz to prevent visible scan lines) -CPU-Z benchmark and stress test used to verify results and temps under load -“Hardware Info” app(CPU-Z company makes this) -with my current Overclock and thermal Grizzly contact frame Cinebench r23 seems to crash when starting the multi-core test(pc games are stable, but perhaps the thermal Grizzly contact frame mounting pressure is causing some stability issues with my ram stability). Question: with the Thermalright LGA1700 contact frame, did you screw down the screws until they essentially came to a stop or did you apply less pressure than this? I’ve heard from other owners of this contact frame that you may need to tighten it down to you feel a stop and you can’t really screw it down anymore. This is the total opposite of the Thermal Grizzly contact frame installation of only going 1/4 turn past lightly screwing down the screws with just your thumb and index finger around the torx screwdriver. Also, some people have realized that their stock retention bracket screws were a little too short to properly reach the backplate when using a contact frame and forcing it to reach with extra pressure just makes it where you can’t fine tune the pressure level of the contact frame since the short screws will become tight after just getting them barely into the backplate holes, which makes it impossible to judge how much pressure is being applied(likely way too much in my case). I did take out another set of screws from an older z790 motherboard in hopes that the screws are longer or the backplate is slightly thinner. I was using an MSI Meg Z790 Ace motherboard with my 13900ks, but it had a tendency to run much hotter for some reason with the way it applied voltage and power compared to my current Asus Z790 Apex motherboard. With the MSI I would reach 95-100c in some benchmarks even with thermal velocity boost in place where it down binned the frequency at certain temperature points set. With the Asus Z790 Apex, I’ve never experienced anything above 81C max in stress tests.
Hi, Thanks for watching. Thank you for sharing such a detailed account of your findings and it is really interesting to see the high temps with the MSI board. So as far as tightening the screws is concerned, I kept going 1/4 at a time till it was very slightly tight, about the same tightness as original ILM screws and I did have to tighten it again slightly as I had crashes. Did you install the thermalright yet? any improvements?
@@TKsTechTalk installed the Thermalright contact frame and my system ram seems to be more stable now at 7800mhz than before. Less game crashes. However I have found that if I set the “when using all 8 P cores” to 5.9ghz it becomes unstable in some games, where as 5.8ghz on 8 cores active is very stable. This is with SVID behavior on “trained” as using typical or worse case scenarios use higher core voltage and higher temps. Now the typical scenario may provide more stability at 5.9ghz. I need to try it out tonight when I get home and see what my temps are with all P cores synced to 5.9Ghz. I typically keep E cores at 4.3ghz as most gaming performance comes from the P cores. I noticed that Battlefield 2042 uses all 8 P cores at one time and is a good game to test system overclock stability. One question I have, is it normal for the CPU Package to be around 7 C lower than the internal CPU temp? My motherboards Q LED seems to report just the cpu package as Hardware Info reports the internal CPU temp and it’s always higher than the motherboards display readout. Thank you!
@@hadleys.4869 Thanks for the detailed reply again. This is interesting seems you are getting good results might have to try some overclocking myself with help from your findings. So as for the package temp, my readout usually reads about the same as the hottest core so i'm not sure what it actually refers to i'm not too sure about that, but its also interesting that you have differences in temps from motherboard and hwinfo. Either way seems you have a great setup! Enjoy :).
I noticed in your HWMon, you have CPU and PUMP fan speeds. The Liquid Freezer II 420 RGB has one header for the "CPU_FAN" header and the other for RGB. How did you get the PUMP fan speed reported with the 420, or is that something else connected? I have the exact same set up and I'm getting 36k Cinebench scores, hitting around 280w, peak at 312w sometimes. I think there is an issue with my mounting because my thermal paste looks nothing like yours, there are spider cracks in it and it's almost watery (Noctua paste). BTW, I get to 100 degrees after 5 seconds of running Cinebench, even on a cold boot. Idle temp is about 36-42. Also, when you installed the 420 rear MB bracket, did the nuts slide into the hole completely and the rear rubber washer compress a little when you tightened the top double-sided screws?
Hi, Thanks for watching and sorry for the delayed response. So for the fan speed, if I remember correctly the CPU fan header actually gives the pump speed and I have the fans connected separately to another header. For mounting I actually just followed the arctic tutorial and though I do remember the back plate touching the CPU back plate I'm sure everything is tightly fitted though I'm guessing this may vary slightly depending on motherboard, though I'm not sure as I haven't tested any other boards. So I believe the rubber would have compressed slightly though I do not recall properly now, if I ever reseat it will let you know. Your temps definitely seem strange and I'm guessing you removed the plastic which is generally a cause of bad contact. How does the thermal paste spread look? I know its been a while so were you able to fix the problem? would be good to know what the solution was may help others too.
I have the same contact frame and Liquid freezer 2 240mm aRGB but at higher ram speeds i get freezes and reboots. Installed stock ilm again and no issues.
Hello I'm trying to decide between a arctic 360 0r 420 for a 13700k right now and im just unsure really. Do you know how clpse the 360 performs at higher wattages you showed in the video with good paste and contact frame? Its very difficult to find this info over 200w.
Hi, Thanks for watching. That's a good question unfortunately I wasn't able to record the installation but I have another video of the 280mm cooler where I show the included LGA1700 kit at the link below: ruclips.net/video/PoFaTM-OEoA/видео.html I essentially just used Arctics youtube video as guide they show all the parts to use, link is below: ruclips.net/video/ZLPDPom8eVQ/видео.html Hope that helps, let us know how you get on. All the best, TK :).
Thanks....very informative video......How do you think this will affect the i7 13700K...Someone is building a workstation for me and they keep insisting that the i9 13900K is much better, but since I will be using Maya for modelling and animation...I believe that Maya uses mainly single core and rendering etc is mostly done on the GPU.....but whatever, I want my CPU to be running as cool as possible..at all times
Hi, Thanks for watching. So I don't know much about Maya unfortunately but assuming single core performance if we just go by single core scores in Geekbench the 13900k is about 3-5% faster then 13700k but the cost difference is about 20-25% more than 13700k. The 13700k should in theory be cooler than the 13900k as well due to lower TDP which I'm guessing won't be a massive concern for single thread applications. But if you will mainly be using single core maybe stick to the 13700K for now, I say this because it will cost you less for now and if the rumours are true about 14th gen (or raptor lake refresh) using the same LGA1700 socket you would have an upgrade path to something like the 14900k (if that's what they end up calling it). Alternatively if you don't plan to upgrade anytime soon and go with 13900k now, for single core I just did a quick test and with -0.05v undervolt and Arctic 420 I am able to maintain a temp of less than 65c in CineBench R23, this will vary based on cooling options and the famous silicon lottery but as you can see it should be fine in terms of temp for single core loads. Hope that helps, let us know what you end up choosing and how it turns out, all the best, TK :).
@@TKsTechTalk Wow..thank you for your in depth reply....in your opinion, is there such a thing as ..too cool ?...I've decided to go with the 13700K because I don't think I'll need more than that can give me....I'm getting a 20GB NVIDIA RTX A4500 which will do the rendering ...also the Arctic freeze 420 and a very large case...Fractal Design Meshify 2xl..with as many fans as I can pack in. A friend did bring up that some things have an 'optimal' temp at which they will work best...I just don't know if that would apply in the case of CPU's
@@eggsclusive12 To the best of my knowledge I don't think there's any optimal CPU temperature and I doubt there are any issues with low temps. Take for example extreme overclocking where Liquid Nitrogen is used, it's normal to see temps well below -40c and even -100c, some people even run stress tests at negative temps. What you will see in normal usage and regular cooling solutions idle temp will be around 30c to maybe 40c depending on CPU and also whether you are using liquid or air cooling. Under load Intel and AMD claim CPU run fine upto 95-100c and warrantied for that and they thermal throttle anyway so I think you will not really need to worry about optimal temps. Off the cooling topic but I like your choice of case, Meshify 2 XL is great if you have space and provides support for lots of cooling options and excellent storage. Sounds like a great build, enjoy :) and if you get a chance would be great to know how thermals hold up but with arctic 420 on 13700k I think you will be fine.
I had not good experience with 3 contact frames. They all work good in first 1 -2 months but then temperature start to grow much higher later. So today 1 core reach 97 c while other cores only 79c. I'm sure it is thermal paste not balanced because of contact frame. Or arctic freezer II 420 not good with contact frame.
Hi, thanks for watching. Personally I did not experience this but I can only assume if you are having to remount or tighten screws if they become lose, try tightening just a tiny bit more to see if it holds out longer, if that doesn't help not sure what the issue could be, let us know how you get on might help others as well. Hope that helps all the best, TK :).
@@yooanto9465 Yeah its difficult to know how tight is too tight, but just try to go with thumb and finger till you feel slight resistance then go just a tiny bit more, if you already tried this not sure what it could be. Once you adjust do a stress test with cinebench or something and some ram tests to make sure its stable. Hope you can get it to work better.
0OK! let's assume i made it in 253/253 watts PL1/PL2 performance mode in bios(Just main configuration) And tempereatures will not up to more than 90 Celsius then Question is will it be stable at hard work loadouts like in Unreal Engine Works??
hi i have a system like: i5-13600k rtx 3060 ti 8gb oc is cooler (ak620 deep cool) enought for high preocess or i should go for 360ex argb deep cool? what u suggest urself?
Hi, Thanks for watching. I did some tests with AK620 in a couple of videos which I will link below using a 12900k. What I found is that when limiting the 12900k to 241w and -0.05v undervolt, I hit max temps of 96C and it did not hit 100C or throttle very much. Considering the TDP 13600k tops out at about 181W, I think it should be ok to use an AK620 without thermal throttling. the 360ex AIO might drop a few degrees as I saw with Arctic 280mm but I think the Arctic 280mm is a better cooler than deepcool 360ex. Personally if you will mainly be gaming you won't see such high CPU usage or power draw and the AK620 should the job and will be most cost effective for your setup. If you are comfortable using a contact frame, you might save another couple of degrees, I have another video on that as well, all links below: AK620 3 fans vs 2 fans: ruclips.net/video/AMCwH1LRoZ4/видео.html AK620 vs Arctic 280mm ruclips.net/video/PoFaTM-OEoA/видео.html&t 12900k and 13900k with contact frame: ruclips.net/video/XlUt5wIUJIo/видео.html Hope that helps, feel free to ask any other questions, all the best, TK :).
Hi, Thanks for watching. I have a Phanteks P600s, I believe inside it is the same as a Phanteks P500a which is usually cheaper and fits the same 420mm AIO. Hope that helps, all the best, TK :).
Hi, Thanks for watching. That's really strange, what sort of temps do you get without the frame? If we know temps without the frame we will at least know if the mounting pressure is ok. I'm assuming you already checked but make sure the plastic is fully removed from the cold plate of the AIO. Let us know how you get on, all the best, TK :).
@@alperdemir8574 I did a quick test with no power limits or undervolt and I get an average during the run of about 1.338v and spikes at max of 1.455v (with one core randomly spiking momentarily to 1.464v). When I apply 0.05v undervolt I get average of about 1.280v. And jUst to add after about 8 or 9 passes I was at 99C, 1C higher than my video. What do you see?
I'm gonna drop a gem on everyone. Turn Multi-Core Enhancement off, there is no fps gain and I don't see any performance gain in anything with it. What I do see is 350+ watts pulled for no reason. On Z690 Aorus Master I run 13900k with MCE off, and undervolt of 1.195v (V-Core) and it reaches 262 watts max in Aida64 100% load. Aorus Master knows how to upvolt to 1.32-1.34v when needed, but MCE needs to be off to remain in thermal safety.
Hi, Thanks for watching. I really appreciate your time in sharing this, I must admit have not played with this setting and will definitely try it to see how well it performs. Thanks again, all the best, TK :).
Got a 360mm arctic cooler and a 14900k and i am getting better temps then you when doing benchmarks. Dont know the reason. I do have my pc in a basement meaning i have a cooler room. Btw i have corsair 4000d airflow. Let me know if you have a theory why this may be
Hi, Thanks for watching. That's a great question and I do not really know the answer to this I'm afraid but let's try to apply some logic which might help us understand if this will work. So, if we assume that at idle the CPU power is only a max of 30w (though it's usually lower than this) and we keep the PC running 24/7 this should be fine and its no where near the rated TDP of 241w. I actually have a dual xeon cpu desktop workstation which is on 24/7 and usually low power usage, 3 years later still on and running. I guess the next important thing is, as long as we do not alter voltage to attain the 105w and only limit the power to 105w to keep the CPU running around the desired voltage, in theory this should not affect the CPU in the long run. There is a great video of Long Term testing by Der Bauer in the link below where he does 4000+ hours of continuous testing, I believe there is some OC applied in this case and there are varying results as to how it affects the CPU in the long run. ruclips.net/video/ZAww0c2m-ks/видео.html Sorry I couldn't give you a definite answer as I do not have enough experience in this sort of usage. If you end up trying this or find something useful, please feel free to share your findings, I'd be interested to know and I'm sure it will help others in a similar situation. All the best, TK :).
Amazing video! What would you say is the best aio thermal wise. I don’t care about the looks whatsoever, only the pump quality and rpm which effects cooling performance, I’m not nearly advanced enough for a custom loop so.. also money is not a problem, I could care less about the price,😅🫶
Hi, Thanks for watching. Unfortunately I'm not going to be able to answer this question and I do not want to be biased as I only have two AIO's, which are both Arctic Liquid Freezer II ARGB 280mm and 420mm and I have videos on both. As you can imagine I did some research before buying and these were hard to get hold of at the time, at least in the UK. However there are a few things to think about when purchasing an AIO and I will list them below in no particular order: 1. Which CPU do you own or plan to buy? If it is an i9 or Ryzen 9 especially the newer ones, you will definitely need an AIO with good performance (obviously :)) 2. Which case do you plan to use the AIO in? This is important as you will see many reviews and videos recommending top mounting AIO with exhaust fans and this will also increase life of the AIO. The issue with this is, if you plan to get a massive 420mm AIO like Arctic with a thick radiator cases have to be large and if space is an issue this will need to be considered. 3. Size doesn't always represent best performance, for example you will see that a 280mm AIO can outperform a 360mm AIO and results will vary across manufacturers. 4. Ensure fans on radiators are optimised for static pressure as they are better designed to move air through them. 5. Case fans should be setup in a way where you have more air coming in than out (I think this is called positive pressure) though my wording is not scientific I hope it makes sense :). For case fans I think it's better to consider fans with high air flow rating not high static pressure. Again the case is important here as you will need to check case fan support. 6. Keep a decent air cooler as backup lol, my AIO hasn't failed in the last 12 months and I think they are generally good nowadays but just in case something goes wrong you will have minimum downtime. Sorry I couldn't give you a more definitive answer but I personally lack experience with AIO's and only started using them in about the last 18 months and also not confident with custom loops ... maybe one day I will venture into this. Either way hope this helps, let us know what you end up choosing and how you get on. All the best, TK :).
@@TKsTechTalk Thanks for the detailed response😃 I really appreciate it Tech, currently I have a I9 14900k with a Corsair 7000d so it can fit a 420mm rad on the front or top. I also have like 14 case fans setup in a push pull configuration with 2 external desk fans blowing into the intake. I got 6 noctua NF A-14 iPPC 3000rpm fans for a push pull config for the aio I’m looking to buy(I don’t know what aio I’m going to buy yet) I do alot of overclocking and am planning on pushing it to 6ghz all p core ratio but im not sure if that’s possible without a custom loop, so I’m just looking for the absolute best aio for thermal performance. I appreciate your feedback as always!🙂🫶
Thanks for the video. I'm assembling a 14900k with contact frame and artic liquid freezer with 6 fans too !
420 or 360?
@@alexangell903 420 . I bought a lian li dynamic evo xl to mount it
420 I bought a Lian Li Dynamic Evo XL to fit it@@alexangell903
Hi, Thanks for watching, glad you found it useful. Been wandering how the results differ for 14900k if you are able to please share your results would be great to know. All the best with the build, TK :).
sure thanks to you@@TKsTechTalk
Thanks. I just got my ALF and already have the contact frame installed, the Thermal Grizzly one. Was kind of nervous on mounting it with the all the pressure and such.
Hi, Thanks for watching. You are welcome glad you found it useful. To be honest I am in the same boat, don't normally attempt such mods but had to be attempted with the temps these CPU's attain. Just want to mention that I did actually have an issue when I first mounted it with the 13900k, and had to tighten the screws a little more, was scary that it crashed at first but has been stable ever since. Glad you managed to install it, which ALF did you get and which CPU do use, would be great to know what sort of temps you get if you have time. All the best, TK :).
@@TKsTechTalk I got the Arctic Liquid Freezer II 360, non RGB and the I7 13700K. I haven’t installed the new AIO yet, as it arrived missing hardware, few O-Rings. I’m currently using the 240 H100i Elite AIO from Corsair and the temps bottleneck when stress testing, so I had to undervolt the CPU, which helps for now.
@@haies09 Ok nice, that would be similar in TDP to 12900k but again the IHS seems to be improved on the 13th gen CPU's so I guess you should get decent results but maybe not as big a drop as 12th gen, and undervolting is def the way to go even with the AIO as you may have seen from my tests, without any real performance loss at all. All the best with the AIO upgrade :).
Thank you for this video. I currently have the Thermal Grizzly contact frame, but I feel like I have it too tight, because I’ve experienced some game crashes sporadically with a very light overclock and extremely good temps. However, I think my stock retention bracket screws are a tad too short for the thickness of this new motherboard and it doesn’t allow me to back the screws out to a lighter pressure without pulling the screws out of the backplate. I ordered the Thermalright LGA1700 contact frame yesterday and should have it by this afternoon hopefully so I can swap it out tonight and hopefully have more system/ram stability. My power limits have been removed and I have cores 1-5 at 5.9Ghz and cores 6-8 at 5.8Ghz. Looks to be around 1.43v at the worst(@ idle with windows 11 power plan set to high performance mode, which keeps the processor at 100% power at idle). My temps have never reached over about 75-80c in all the milder benchmarks(CPU-Z multi-core benchmark(71-75C) and stress test(80-81C) functions) I’ve done with an idle temp of around 26-30C depending on room temp/AC on. I have “SVID Behavior” set to “Trained Behavior” instead of Auto or Typical behavior. This helps keep the voltage and temps down compared to using “typical” or “worst case scenario”(intel’s model, which can pump up to 1.8v!!!)”, which will pump in way more voltage than needed and increase the temps. I set Load Line Calibration to level 4(Asus) to reduce Vdroop a little, which improves OC stability. I may bump this to level 6. Not sure yet. My system specs are as follows:
-Intel 13900ks CPU
-Asus ROG Maximus Z790 APEX (2-DIMM motherboard)
-G.Skill 32GB 7800mhz XMP DDR5 kit
-Corsair h170i elite lcd 420mm AIO(6 push/pull fans pulling in cool air)
-Thermal Grizzly 12/13th gen contact frame(soon to be Thermalright version)
-Samsung 990 Pro 2TB nvme SSD
-Gigabyte Gaming OC 4090(stock 450watts)
-Asus Thor 1600watt PSU
-Corsair 1000D Super tower case(tons of room, but very heavy)
-Windows 11 Home
-Samsung Odyssey Neo G8 4k 240Hz monitor(I reduced this to 120Hz to prevent visible scan lines)
-CPU-Z benchmark and stress test used to verify results and temps under load
-“Hardware Info” app(CPU-Z company makes this)
-with my current Overclock and thermal Grizzly contact frame Cinebench r23 seems to crash when starting the multi-core test(pc games are stable, but perhaps the thermal Grizzly contact frame mounting pressure is causing some stability issues with my ram stability).
Question: with the Thermalright LGA1700 contact frame, did you screw down the screws until they essentially came to a stop or did you apply less pressure than this? I’ve heard from other owners of this contact frame that you may need to tighten it down to you feel a stop and you can’t really screw it down anymore. This is the total opposite of the Thermal Grizzly contact frame installation of only going 1/4 turn past lightly screwing down the screws with just your thumb and index finger around the torx screwdriver. Also, some people have realized that their stock retention bracket screws were a little too short to properly reach the backplate when using a contact frame and forcing it to reach with extra pressure just makes it where you can’t fine tune the pressure level of the contact frame since the short screws will become tight after just getting them barely into the backplate holes, which makes it impossible to judge how much pressure is being applied(likely way too much in my case). I did take out another set of screws from an older z790 motherboard in hopes that the screws are longer or the backplate is slightly thinner. I was using an MSI Meg Z790 Ace motherboard with my 13900ks, but it had a tendency to run much hotter for some reason with the way it applied voltage and power compared to my current Asus Z790 Apex motherboard. With the MSI I would reach 95-100c in some benchmarks even with thermal velocity boost in place where it down binned the frequency at certain temperature points set. With the Asus Z790 Apex, I’ve never experienced anything above 81C max in stress tests.
Hi, Thanks for watching. Thank you for sharing such a detailed account of your findings and it is really interesting to see the high temps with the MSI board. So as far as tightening the screws is concerned, I kept going 1/4 at a time till it was very slightly tight, about the same tightness as original ILM screws and I did have to tighten it again slightly as I had crashes. Did you install the thermalright yet? any improvements?
@@TKsTechTalk installed the Thermalright contact frame and my system ram seems to be more stable now at 7800mhz than before. Less game crashes. However I have found that if I set the “when using all 8 P cores” to 5.9ghz it becomes unstable in some games, where as 5.8ghz on 8 cores active is very stable. This is with SVID behavior on “trained” as using typical or worse case scenarios use higher core voltage and higher temps. Now the typical scenario may provide more stability at 5.9ghz. I need to try it out tonight when I get home and see what my temps are with all P cores synced to 5.9Ghz. I typically keep E cores at 4.3ghz as most gaming performance comes from the P cores. I noticed that Battlefield 2042 uses all 8 P cores at one time and is a good game to test system overclock stability. One question I have, is it normal for the CPU Package to be around 7 C lower than the internal CPU temp? My motherboards Q LED seems to report just the cpu package as Hardware Info reports the internal CPU temp and it’s always higher than the motherboards display readout. Thank you!
@@hadleys.4869 Thanks for the detailed reply again. This is interesting seems you are getting good results might have to try some overclocking myself with help from your findings. So as for the package temp, my readout usually reads about the same as the hottest core so i'm not sure what it actually refers to i'm not too sure about that, but its also interesting that you have differences in temps from motherboard and hwinfo. Either way seems you have a great setup! Enjoy :).
I noticed in your HWMon, you have CPU and PUMP fan speeds. The Liquid Freezer II 420 RGB has one header for the "CPU_FAN" header and the other for RGB. How did you get the PUMP fan speed reported with the 420, or is that something else connected? I have the exact same set up and I'm getting 36k Cinebench scores, hitting around 280w, peak at 312w sometimes. I think there is an issue with my mounting because my thermal paste looks nothing like yours, there are spider cracks in it and it's almost watery (Noctua paste). BTW, I get to 100 degrees after 5 seconds of running Cinebench, even on a cold boot. Idle temp is about 36-42. Also, when you installed the 420 rear MB bracket, did the nuts slide into the hole completely and the rear rubber washer compress a little when you tightened the top double-sided screws?
Hi, Thanks for watching and sorry for the delayed response.
So for the fan speed, if I remember correctly the CPU fan header actually gives the pump speed and I have the fans connected separately to another header.
For mounting I actually just followed the arctic tutorial and though I do remember the back plate touching the CPU back plate I'm sure everything is tightly fitted though I'm guessing this may vary slightly depending on motherboard, though I'm not sure as I haven't tested any other boards. So I believe the rubber would have compressed slightly though I do not recall properly now, if I ever reseat it will let you know.
Your temps definitely seem strange and I'm guessing you removed the plastic which is generally a cause of bad contact. How does the thermal paste spread look? I know its been a while so were you able to fix the problem? would be good to know what the solution was may help others too.
I have the same contact frame and Liquid freezer 2 240mm aRGB but at higher ram speeds i get freezes and reboots. Installed stock ilm again and no issues.
Hello I'm trying to decide between a arctic 360 0r 420 for a 13700k right now and im just unsure really. Do you know how clpse the 360 performs at higher wattages you showed in the video with good paste and contact frame? Its very difficult to find this info over 200w.
Which stand off you used for arctic freezer aio?
I'm confused i got very short , medium and long screws?
Hi, Thanks for watching. That's a good question unfortunately I wasn't able to record the installation but I have another video of the 280mm cooler where I show the included LGA1700 kit at the link below:
ruclips.net/video/PoFaTM-OEoA/видео.html
I essentially just used Arctics youtube video as guide they show all the parts to use, link is below:
ruclips.net/video/ZLPDPom8eVQ/видео.html
Hope that helps, let us know how you get on. All the best, TK :).
Where did you set your 241 watt power limit and .05 undervolt?
Nice!!!!
Hi, Thanks for watching, glad you found it useful.
I have the same AIO but I also have 95 degrees with 250W. I don't know why but you have a much better results than me
Thanks....very informative video......How do you think this will affect the i7 13700K...Someone is building a workstation for me and they keep insisting that the i9 13900K is much better, but since I will be using Maya for modelling and animation...I believe that Maya uses mainly single core and rendering etc is mostly done on the GPU.....but whatever, I want my CPU to be running as cool as possible..at all times
Hi, Thanks for watching. So I don't know much about Maya unfortunately but assuming single core performance if we just go by single core scores in Geekbench the 13900k is about 3-5% faster then 13700k but the cost difference is about 20-25% more than 13700k. The 13700k should in theory be cooler than the 13900k as well due to lower TDP which I'm guessing won't be a massive concern for single thread applications.
But if you will mainly be using single core maybe stick to the 13700K for now, I say this because it will cost you less for now and if the rumours are true about 14th gen (or raptor lake refresh) using the same LGA1700 socket you would have an upgrade path to something like the 14900k (if that's what they end up calling it).
Alternatively if you don't plan to upgrade anytime soon and go with 13900k now, for single core I just did a quick test and with -0.05v undervolt and Arctic 420 I am able to maintain a temp of less than 65c in CineBench R23, this will vary based on cooling options and the famous silicon lottery but as you can see it should be fine in terms of temp for single core loads.
Hope that helps, let us know what you end up choosing and how it turns out, all the best, TK :).
@@TKsTechTalk Wow..thank you for your in depth reply....in your opinion, is there such a thing as ..too cool ?...I've decided to go with the 13700K because I don't think I'll need more than that can give me....I'm getting a 20GB NVIDIA RTX A4500 which will do the rendering ...also the Arctic freeze 420 and a very large case...Fractal Design Meshify 2xl..with as many fans as I can pack in. A friend did bring up that some things have an 'optimal' temp at which they will work best...I just don't know if that would apply in the case of CPU's
@@eggsclusive12 To the best of my knowledge I don't think there's any optimal CPU temperature and I doubt there are any issues with low temps. Take for example extreme overclocking where Liquid Nitrogen is used, it's normal to see temps well below -40c and even -100c, some people even run stress tests at negative temps.
What you will see in normal usage and regular cooling solutions idle temp will be around 30c to maybe 40c depending on CPU and also whether you are using liquid or air cooling. Under load Intel and AMD claim CPU run fine upto 95-100c and warrantied for that and they thermal throttle anyway so I think you will not really need to worry about optimal temps.
Off the cooling topic but I like your choice of case, Meshify 2 XL is great if you have space and provides support for lots of cooling options and excellent storage. Sounds like a great build, enjoy :) and if you get a chance would be great to know how thermals hold up but with arctic 420 on 13700k I think you will be fine.
@@TKsTechTalk Thanks...I will update when I get the PC.....and thank you again for taking the time to reply
@@eggsclusive12 you are most welcome :)
I had not good experience with 3 contact frames. They all work good in first 1 -2 months but then temperature start to grow much higher later. So today 1 core reach 97 c while other cores only 79c. I'm sure it is thermal paste not balanced because of contact frame. Or arctic freezer II 420 not good with contact frame.
Hi, thanks for watching. Personally I did not experience this but I can only assume if you are having to remount or tighten screws if they become lose, try tightening just a tiny bit more to see if it holds out longer, if that doesn't help not sure what the issue could be, let us know how you get on might help others as well. Hope that helps all the best, TK :).
@@TKsTechTalk thanks can I ask you when to stop tighten it? It won't damage the motherboard
@@yooanto9465 Yeah its difficult to know how tight is too tight, but just try to go with thumb and finger till you feel slight resistance then go just a tiny bit more, if you already tried this not sure what it could be. Once you adjust do a stress test with cinebench or something and some ram tests to make sure its stable. Hope you can get it to work better.
0OK! let's assume i made it in 253/253 watts PL1/PL2 performance mode in bios(Just main configuration) And tempereatures will not up to more than 90 Celsius then Question is will it be stable at hard work loadouts like in Unreal Engine Works??
hi i have a system like:
i5-13600k
rtx 3060 ti 8gb oc
is cooler (ak620 deep cool) enought for high preocess or i should go for 360ex argb deep cool?
what u suggest urself?
Hi, Thanks for watching. I did some tests with AK620 in a couple of videos which I will link below using a 12900k. What I found is that when limiting the 12900k to 241w and -0.05v undervolt, I hit max temps of 96C and it did not hit 100C or throttle very much. Considering the TDP 13600k tops out at about 181W, I think it should be ok to use an AK620 without thermal throttling. the 360ex AIO might drop a few degrees as I saw with Arctic 280mm but I think the Arctic 280mm is a better cooler than deepcool 360ex. Personally if you will mainly be gaming you won't see such high CPU usage or power draw and the AK620 should the job and will be most cost effective for your setup. If you are comfortable using a contact frame, you might save another couple of degrees, I have another video on that as well, all links below:
AK620 3 fans vs 2 fans:
ruclips.net/video/AMCwH1LRoZ4/видео.html
AK620 vs Arctic 280mm
ruclips.net/video/PoFaTM-OEoA/видео.html&t
12900k and 13900k with contact frame:
ruclips.net/video/XlUt5wIUJIo/видео.html
Hope that helps, feel free to ask any other questions, all the best, TK :).
radyator turn fan direction. no insede, out side
what case do you use?
Hi, Thanks for watching. I have a Phanteks P600s, I believe inside it is the same as a Phanteks P500a which is usually cheaper and fits the same 420mm AIO. Hope that helps, all the best, TK :).
Same aio, same cpu, same wattage, same bracket but mine is reaching 100C after 3 secs R23 started. My idle temp is 35C. Whats wrong with me ?
Hi, Thanks for watching. That's really strange, what sort of temps do you get without the frame? If we know temps without the frame we will at least know if the mounting pressure is ok. I'm assuming you already checked but make sure the plastic is fully removed from the cold plate of the AIO. Let us know how you get on, all the best, TK :).
@@TKsTechTalkwhat is your vcore while running r23 (13900K)
@@alperdemir8574 I did a quick test with no power limits or undervolt and I get an average during the run of about 1.338v and spikes at max of 1.455v (with one core randomly spiking momentarily to 1.464v). When I apply 0.05v undervolt I get average of about 1.280v. And jUst to add after about 8 or 9 passes I was at 99C, 1C higher than my video. What do you see?
I'm gonna drop a gem on everyone. Turn Multi-Core Enhancement off, there is no fps gain and I don't see any performance gain in anything with it.
What I do see is 350+ watts pulled for no reason.
On Z690 Aorus Master I run 13900k with MCE off, and undervolt of 1.195v (V-Core) and it reaches 262 watts max in Aida64 100% load.
Aorus Master knows how to upvolt to 1.32-1.34v when needed, but MCE needs to be off to remain in thermal safety.
Hi, Thanks for watching. I really appreciate your time in sharing this, I must admit have not played with this setting and will definitely try it to see how well it performs. Thanks again, all the best, TK :).
Got a 360mm arctic cooler and a 14900k and i am getting better temps then you when doing benchmarks. Dont know the reason. I do have my pc in a basement meaning i have a cooler room. Btw i have corsair 4000d airflow. Let me know if you have a theory why this may be
Whats your opinion on keeping 13900k on 105W whole time? Ill be doing 24/7 render and I wanna lower the temps and power usage as much as possible.
Hi, Thanks for watching. That's a great question and I do not really know the answer to this I'm afraid but let's try to apply some logic which might help us understand if this will work. So, if we assume that at idle the CPU power is only a max of 30w (though it's usually lower than this) and we keep the PC running 24/7 this should be fine and its no where near the rated TDP of 241w. I actually have a dual xeon cpu desktop workstation which is on 24/7 and usually low power usage, 3 years later still on and running.
I guess the next important thing is, as long as we do not alter voltage to attain the 105w and only limit the power to 105w to keep the CPU running around the desired voltage, in theory this should not affect the CPU in the long run. There is a great video of Long Term testing by Der Bauer in the link below where he does 4000+ hours of continuous testing, I believe there is some OC applied in this case and there are varying results as to how it affects the CPU in the long run.
ruclips.net/video/ZAww0c2m-ks/видео.html
Sorry I couldn't give you a definite answer as I do not have enough experience in this sort of usage. If you end up trying this or find something useful, please feel free to share your findings, I'd be interested to know and I'm sure it will help others in a similar situation. All the best, TK :).
@@TKsTechTalk thanks a lot!
@@skrajina8037 you are most welcome :), hope it was of some use.
Amazing video! What would you say is the best aio thermal wise. I don’t care about the looks whatsoever, only the pump quality and rpm which effects cooling performance, I’m not nearly advanced enough for a custom loop so.. also money is not a problem, I could care less about the price,😅🫶
Hi, Thanks for watching. Unfortunately I'm not going to be able to answer this question and I do not want to be biased as I only have two AIO's, which are both Arctic Liquid Freezer II ARGB 280mm and 420mm and I have videos on both. As you can imagine I did some research before buying and these were hard to get hold of at the time, at least in the UK.
However there are a few things to think about when purchasing an AIO and I will list them below in no particular order:
1. Which CPU do you own or plan to buy? If it is an i9 or Ryzen 9 especially the newer ones, you will definitely need an AIO with good performance (obviously :))
2. Which case do you plan to use the AIO in? This is important as you will see many reviews and videos recommending top mounting AIO with exhaust fans and this will also increase life of the AIO. The issue with this is, if you plan to get a massive 420mm AIO like Arctic with a thick radiator cases have to be large and if space is an issue this will need to be considered.
3. Size doesn't always represent best performance, for example you will see that a 280mm AIO can outperform a 360mm AIO and results will vary across manufacturers.
4. Ensure fans on radiators are optimised for static pressure as they are better designed to move air through them.
5. Case fans should be setup in a way where you have more air coming in than out (I think this is called positive pressure) though my wording is not scientific I hope it makes sense :). For case fans I think it's better to consider fans with high air flow rating not high static pressure. Again the case is important here as you will need to check case fan support.
6. Keep a decent air cooler as backup lol, my AIO hasn't failed in the last 12 months and I think they are generally good nowadays but just in case something goes wrong you will have minimum downtime.
Sorry I couldn't give you a more definitive answer but I personally lack experience with AIO's and only started using them in about the last 18 months and also not confident with custom loops ... maybe one day I will venture into this.
Either way hope this helps, let us know what you end up choosing and how you get on. All the best, TK :).
@@TKsTechTalk Thanks for the detailed response😃 I really appreciate it Tech, currently I have a I9 14900k with a Corsair 7000d so it can fit a 420mm rad on the front or top. I also have like 14 case fans setup in a push pull configuration with 2 external desk fans blowing into the intake. I got 6 noctua NF A-14 iPPC 3000rpm fans for a push pull config for the aio I’m looking to buy(I don’t know what aio I’m going to buy yet) I do alot of overclocking and am planning on pushing it to 6ghz all p core ratio but im not sure if that’s possible without a custom loop, so I’m just looking for the absolute best aio for thermal performance. I appreciate your feedback as always!🙂🫶