Caporegime
To be fair the difference between 5.8Ghz (13900K) and 6.0Ghz (13900KS) is 3%, he's getting just under 2% performance gain, seems about right.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
13900KS review
- Thread starter Grim5
- Start date
To be fair the difference between 5.8Ghz (13900K) and 6.0Ghz (13900KS) is 3%, he's getting just under 2% performance gain, seems about right.
Caporegime
Wait.... "small air cooler / 330 watts" what's this small air cooler, NH-D15?
Soldato
U12a
Caporegime
That is what i would call a mid size air cooler, what sort of temps are you getting at 330 watts?
Last edited:
Soldato
For 24/7 im using an 85c temp limit / no power limit and it hits around 280 watts at 85c. For benchmarks i use a 95c temp limit and it hits 330 to 350 watts at 95c in cbr23 and ycruncher.
Tried the frame, didn't notice any kind of difference so i removed it.
For comparison, the 12900k is hitting 100c at around 270 watts with same cooler, that's what i mean when i say the 13900k is much easier to tame.
EG1. Was getting better temps with the z690 apex but i rmaed it, got a unify x now
Last edited:
Caporegime
Nvidia should put the NH-U12A on the 4090 instead of those 3KG monstrosities.
Soldato
Would you be surprised if i told you the u12a is one of the heaviest coolers out there? Its small, but the fin density is really high and it has 7 heatpipes instead of the usual 6 like the d15
Caporegime
Noctua make the best air coolers you can buy, that i do know.
Last edited:
Soldato
Usually yeah, right now the fc140 from thermalright is the got if size is no concern.
Soldato
- Joined
- 28 Sep 2018
- Posts
- 2,533
Steve was also managing to hit 100c on a stock 13600k with a 360mm AIO which is just clowny.
For ref, my average chip at 5.6 all core is running 85c max on something brutal like ycruncher 5b and barely hitting 80c in an hour of occt large/avx2 with a 280mm aio.
As always, there's no competency requirement to having a YT channel.
For ref, my average chip at 5.6 all core is running 85c max on something brutal like ycruncher 5b and barely hitting 80c in an hour of occt large/avx2 with a 280mm aio.
As always, there's no competency requirement to having a YT channel.
Last edited:
Caporegime
Steve didn't make a bad choice with the Arctic Liquid Freezer II 420mm, its the best pre-built cooler money can buy, its better than the NH-D15, the only way to get better cooling is to go with a good custom loop.
I have the 360 version, they have rad's that are more higher end custom loop than AIO, thick chunky crimp fitted pipes and quite a powerful pump, they are NOT Asetek.
I have the 360 version, they have rad's that are more higher end custom loop than AIO, thick chunky crimp fitted pipes and quite a powerful pump, they are NOT Asetek.
Last edited:
Steve was also managing to hit 100c on a stock 13600k with a 360mm AIO which is just clowny.
For ref, my average chip at 5.6 all core is running 85c max on something brutal like ycruncher 5b and barely hitting 80c in an hour of occt large/avx2 with a 280mm aio.
As always, there's no competency requirement to having a YT channel.
It’s summer in Aus right now, so maybe in high ambient temp. Still his temps are awful and even has e cores off in the 13900KS review
Last edited:
Soldato
- Joined
- 28 Sep 2018
- Posts
- 2,533
That’s his stock 13600k temps I posted lol
Soldato
Nobody is questioning how good the LFII is. It's great, we can all agree on that. Im questioning his results with that cooler. Look at this for example, a u12a at 330watts hitting 95c in ycruncher. That's an insanely heavy workload btw, and my ambient was pretty high at the time - around 23 to 24cSteve didn't make a bad choice with the Arctic Liquid Freezer II 420mm, its the best pre-built cooler money can buy, its better than the NH-D15, the only way to get better cooling is to go with a good custom loop.
I have the 360 version, they have rad's that are more higher end custom loop than AIO, thick chunky crimp fitted pipes and quite a powerful pump, they are NOT Asetek.
According to his review (minute 4:12) he is hitting 100c with a 420 LF2 running cbr23 at 280 watts, while also using a contact frame. That's just clowny
Last edited:
Soldato
Actually I kept watching his review, more clowny stuff. At minute 5:03 he is using a pure loop 2 fx and he is hitting 101c at 250w 
I mean come on, unless the bequiet AIO is complete trash, his numbers don't make sense.
I mean come on, unless the bequiet AIO is complete trash, his numbers don't make sense.
Soldato
EKWB just released their direct die water cooling block For the 13900KS. You'll pretty much need it
Direct Die Cooling for the Intel 12th and 13th-Gen Core Series CPUs - ekwb.com
EK, the premium liquid cooling hardware manufacturer, is launching the Direct Die product series aimed at delidding and achieving the ultimate cooling of the LGA 1700 socket-based Intel CPUs from the 12th generation Alder Lake and 13th gen Raptor Lake families.www.ekwb.com
DerBauer just released a video using it. Great temp drops and most likely leave OC headroom.
Caporegime
Guru3D, Toms Hardware and Tech Power Up all hit 100c on the 13900K, TPU actually hit 117c with power limits removed.
HUB aren't seeing anything unusual. the thing uses 300 watts, that's a lot for any cooler to deal with.
HUB aren't seeing anything unusual. the thing uses 300 watts, that's a lot for any cooler to deal with.
Last edited:
Soldato
- Joined
- 28 Sep 2018
- Posts
- 2,533
So I put my 13600k to stock frequencies and ran r23 which is what Steve ran in his 13600k review where he hit tjmax right away. Because he thermal thtrottles right away and gets a frequency drop, it also lowered his score by ~900 points over a true stock 13600k score.
My temps are miles off his. Lets another 10c to mine for difference in ambients and even and he's more than 25c! higher than me. While it may seem like I'm picking on Steve, I'm really not. The techreviewer industry is by the far the most incompetent out of my hobbies. The same applies to the the rest of the mainstream reviewers.
My temps are miles off his. Lets another 10c to mine for difference in ambients and even and he's more than 25c! higher than me. While it may seem like I'm picking on Steve, I'm really not. The techreviewer industry is by the far the most incompetent out of my hobbies. The same applies to the the rest of the mainstream reviewers.
Last edited:
'Traditional' oldschool overclocking has been dead for a while apart from a few niche cases, I wrote about it a while back but everything these days seems to be about doing small tweaks on optimising the 'curves; via whatever black art of tweaks is needed but fundamentally most chips these days are just built to scale automatically and then binned based on whereabout they fall.
The glory days of buying a CPU for a third of the price of the high end part, bumping the bus speed by 50% and tweaking voltage to make it stable are long gone. 20 year old me would be ashamed looking at me now, for the past few years basically running cpus at stock speeds with a few tweaks but perhaps only getting a peak clock 20% higher at most than whatever turbo/boost clock comes out of the box.
Associate
This often seems the case as both my 5900X and 5800X scored around at least 5% better in CB23 than in HWU reviews. Granted, I had faster RAM but having played around with it CB23 isn't particularly sensitive to it, and they used a higher end motherboard too.
It's not going to be a case of high ambient temps in Oz as surely they do the testing in an air conditioned environment. There'll always be variances in testing due to different configurations, software etc. At least in their case it seems quite consistent judging from our experiences comparing to HWU benchmarks.
For the most comprehensive controlled tests Gamers Nexus are right up there so will be interesting to see what they observe with the KS. With the X3D's releasing soon it seems AMD will have the gaming and efficiency crowns, but in the lower SKUs in particular Intel will be the more balanced setups in terms of similar gaming performance but better production capabilities