Coffeelake thread, benchmarks, reviews, input, everything.

Page 40 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
What is the cas settings for 2666? Its c16 for 3200.
My point beeing if its same cas its kind of misleading test. You can eg run the 8400 on a non oc z board on 2666 c14 easily. You can tweak latency and settings not bandwith.
Steve knows that and to be fair it irritates me he just dint flat out write the latency for the 2666 ram but gives a non link to prior cpu reviews. It comes off a bit fishy.

In more general view many games is latency dependant and thats why we see ryzen perform worse. Especially with slow c16 settings. I simply dont agree in comparing platforms like that. Ryzen needs low latency and a 8700k a good cooler so it dont throttle. If you dont give them that its imo biased testing.

In this the purpose seems to be look 3200 is much better for a 8400 go get you 370 but imo it comes of as more as a sales piece. I need some c14 2666 to confirm.
I am pretty sure the results is more or less excactly the same.
Unless proven otherwise, you would have to assume the DDR4 2666 and 3200 are run at identical timings. Yes it would have been good to know the exact timings, why don't you post a comment on their YouTube channel and see if you get an answer?

Accusing Steve of foul play and trying to promote Z370 sales is going a bit far, he is one of the best Youtube reviewers IMO and always seems fair in his appraisal of products. Innocent until proven guilty here, I'm going to give Steve the benefit of the doubt based on his track record.

By the way, Techspot tested the i3 8100 and compared performance between DDR4-2400 and DDR4-3200, and shows similar results to Steve: https://www.techspot.com/review/1499-intel-core-i3-8100-i3-8350K/page3.html

Some games just scale well with increased memory bandwith, it's not just as simple as saying if you run at CL14 regardless of your RAM speed you are good to go. I'm sure there are appreciable differences between DDR4-3200 CL14 and DDR4 2400/2666 CL14 in certain workloads and games.
 
Last edited:

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Interesting.
https://www.youtube.com/watch?v=87M3QdEzRFk

this shows that with faster ram R7 1700 can fight i7 8700K @5GHz 3000MHz DDr4.
Yeah, not really. They don't even test the 8700K in that video, so I'm not sure how you can come to that conclusion. Meanwhile, a site that actually tested both CPUs in The Witcher 3:
http://www.eurogamer.net/articles/digitalfoundry-2017-intel-coffee-lake-core-i7-8700k-review

8700K @ 5GHz DDR4-3000: 172.4fps
1700 @ 4GHz DDR4-3200: 121.3fps

So with slower RAM the 8700K OC runs 42% faster than the 1700 OC in The Witcher 3.
 

IRobot23

Senior member
Jul 3, 2017
601
183
76
Yeah, not really:
http://www.eurogamer.net/articles/digitalfoundry-2017-intel-coffee-lake-core-i7-8700k-review

Witcher 3
8700K @ 5GHz DDR4-3000: 172.4fps
1700 @ 4GHz DDR4-3200: 121.3fps

So with slower RAM the 8700K OC runs 42% faster than the 1700 OC in The Witcher 3.
I did re-edit, since I have mistakenly posted. Techspot shows R5 1600 as fast as i7 7800X while digital foundry will show it contrary.
He did get low fps in AotS, my R7 1700 hits 49-50fps at 3,8GHz with 8GB of ddr4.
 
Last edited:

IRobot23

Senior member
Jul 3, 2017
601
183
76

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
https://www.gamersnexus.net/images/media/2017/CPUs/i5-8400/i5-8400-legacy-wd2.png

What's with strange test ? Intel 4790K is pretty damn close to 8700K Stock.also huge difference between various CPUs.Game engine is Crap? or due to DrawCall back ?
What is wrong with those numbers? A stock 4790K is still a pretty fast CPU for gaming, and the 8700K is about 12% ahead in both avg and mins. Its part of their 'Legacy' tests too, btw. I think in newer games (especially those that can take advantage of >8 threads) the 8700K would be further ahead of the 4790K.

Edit - Another reason the 4790K performs relatively well against the newer CPUs is because it is running DDR3-2400, which is the pinnacle of DDR3 techlology. It's akin to running DDR4-4000 (or faster) on the Kaby/Coffee Lake platforms. So we're looking at a 'best case / maxed out' 4790K versus a 'regular' DDR4-3200 based Kaby/CFL platform
 
Last edited:

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
I did re-edit, since I have mistakenly posted. Techspot shows R5 1600 as fast as i7 7800X while digital foundry will show it contrary.
He did get low fps in AotS, my R7 1700 hits 49-50fps at 3,8GHz with 8GB of ddr4.
I'm not sure what point you're trying to make now? Initially you said the 1700 + faster DDR4 can compete with a 8700K @ 5GHz in gaming, which is not the case as Digital Foundrys review shows. Now you're talking about Ryzen 5 and SkyLake X, seriously your argument jumps around more than a grasshopper! ;)
 

IRobot23

Senior member
Jul 3, 2017
601
183
76
DDR4 3000 starts at 154$.
https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007611 600006072 600561669&IsNodeId=1&bop=And&PageSize=60&order=PRICE

15-16-16-35
https://www.newegg.com/Product/Product.aspx?Item=N82E16820231937

CL 15 15 15 35 @ 160$
https://www.newegg.com/Product/Product.aspx?Item=N82E16820313620

CL 16 16 16 36 @ 3200MT/s @ 165$
https://www.newegg.com/Product/Product.aspx?Item=N82E16820313622

I'm not sure what point you're trying to make now? Initially you said the 1700 + faster DDR4 can compete with a 8700K @ 5GHz in gaming, which is not the case as Digital Foundrys review shows. Now you're talking about Ryzen 5 and SkyLake X, seriously your argument jumps around more than a grasshopper! ;)
Initially I did post before I re-edit my post. Like I said digital foundry shows i7 7800X faster than R5 1600X @ 4GHz while techspot shows different things.

And as Digitalfoundry mentions, more complex screens (more CPU intensive)
https://www.youtube.com/watch?v=lMifuDMrv-Q

Even i7 8700K still at 4,8GHz still cannot run much faster than R7 1700/1600 @ 4GHz and fast ram.
https://www.youtube.com/watch?v=xeAy4RiGVZw&t=56s

Yes, different maps etc. But shows that even 6 core skylake at 4,8GHz cannot handle and maybe in this scenario quad channel will help CPU.

Initially I wanted to say that Digital foundry shows i7 8700K in best light as they did few months ago with i7 7700K.

However, nobody is having any problem with coffee lake? no bugs?...

Definitely i7 8700K is fastest gaming CPU by far from those test, but those competitive games are not even tested here. They do talk about complexity in games and how threads do help, yet they haven tested games...

I did run benchmark yesterday, AotS R5 1600 (mimic R7), 3,8GHZ 3333 CL16 16 16 39 and I score above 44fps in CPU test. I do have only 8GB of DDR4.
 
Last edited:

Dayman1225

Golden Member
Aug 14, 2017
1,081
791
146
"as of now, we are not sure whether B/H boards will support only the native memory speed of 2666MHz or higher multipliers. Some board vendors have suggested a possibility of unlocked memory multipliers of 32/36x, but haven’t confirmed, while other sources have suggested a maximum speed of 2666MHz. Because we cannot reasonably confirm either, we decided to just test both, then let the chips fall where they may in 1Q18."

Source - Gamers Nexus 8400 2666/3200mhz review
 
  • Like
Reactions: Campy

IRobot23

Senior member
Jul 3, 2017
601
183
76
Building i7 8700 system for my friend, can anyone help me with motherboard choice? 140$ is maximum.
 

LTC8K6

Lifer
Mar 10, 2004
28,520
1,574
126
"as of now, we are not sure whether B/H boards will support only the native memory speed of 2666MHz or higher multipliers. Some board vendors have suggested a possibility of unlocked memory multipliers of 32/36x, but haven’t confirmed, while other sources have suggested a maximum speed of 2666MHz. Because we cannot reasonably confirm either, we decided to just test both, then let the chips fall where they may in 1Q18."

Source - Gamers Nexus 8400 2666/3200mhz review
I can't recall a non-Z board ever allowing above stock memory speeds, except for the brief period when we could overclock Haswell chips on certain non-Z boards.
 

Dayman1225

Golden Member
Aug 14, 2017
1,081
791
146
I can't recall a non-Z board ever allowing above stock memory speeds, except for the brief period when we could overclock Haswell chips on certain non-Z boards.
Maybe Intel/Mobo manufactures have had a change of heart since competition appeared. Who knows, we'll have to wait and see.
 

TheLycan

Member
Mar 8, 2017
34
11
36
I can't recall a non-Z board ever allowing above stock memory speeds, except for the brief period when we could overclock Haswell chips on certain non-Z boards.
And some boards with chipset A320 for amd. Still i dont think they will canabalize z370, which are so expensive.
 

TheF34RChannel

Senior member
May 18, 2017
786
309
136
Yeah, not really. They don't even test the 8700K in that video, so I'm not sure how you can come to that conclusion. Meanwhile, a site that actually tested both CPUs in The Witcher 3:
http://www.eurogamer.net/articles/digitalfoundry-2017-intel-coffee-lake-core-i7-8700k-review

8700K @ 5GHz DDR4-3000: 172.4fps
1700 @ 4GHz DDR4-3200: 121.3fps

So with slower RAM the 8700K OC runs 42% faster than the 1700 OC in The Witcher 3.
Imagine higher frequency/lower latencies RAM with that CPU + OC in The Witcher 3 (which repsonds extremely well to faster RAM)... DF did some nice tests with the 7700K + a variety of RAM speed combo's back in the day.

I can't recall a non-Z board ever allowing above stock memory speeds, except for the brief period when we could overclock Haswell chips on certain non-Z boards.
I've had an Asrock P67 board which allowed RAM speeds up to 2133 DDR3 I believe - certainly above stock speeds!
 

krumme

Diamond Member
Oct 9, 2009
5,952
1,584
136
Unless proven otherwise, you would have to assume the DDR4 2666 and 3200 are run at identical timings. Yes it would have been good to know the exact timings, why don't you post a comment on their YouTube channel and see if you get an answer?

Accusing Steve of foul play and trying to promote Z370 sales is going a bit far, he is one of the best Youtube reviewers IMO and always seems fair in his appraisal of products. Innocent until proven guilty here, I'm going to give Steve the benefit of the doubt based on his track record.

By the way, Techspot tested the i3 8100 and compared performance between DDR4-2400 and DDR4-3200, and shows similar results to Steve: https://www.techspot.com/review/1499-intel-core-i3-8100-i3-8350K/page3.html

Some games just scale well with increased memory bandwith, it's not just as simple as saying if you run at CL14 regardless of your RAM speed you are good to go. I'm sure there are appreciable differences between DDR4-3200 CL14 and DDR4 2400/2666 CL14 in certain workloads and games.
Don't bend my point please. 3200c16 runs 2666c14 most of the time. People that get a 8400 can get a non z board and probably use 2666 c14 without any problems vs 3200c16 by just buying 3200c16 and tweak the settings. Steve makes a lot of good stuff but this is nonsense unless he shows it's bandwith. He is promoting something where the numbers miss. I am not saying he is wrong it's just highly doubtful he is imo.
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Don't bend my point please. 3200c16 runs 2666c14 most of the time. People that get a 8400 can get a non z board and probably use 2666 c14 without any problems vs 3200c16 by just buying 3200c16 and tweak the settings. Steve makes a lot of good stuff but this is nonsense unless he shows it's bandwith. He is promoting something where the numbers miss. I am not saying he is wrong it's just highly doubtful he is imo.
Seriously, did you even bother to read the article that goes along with the video? Obviously not, otherwise you wouldn't be making such accusations:

We still don’t have a full picture for this generation of Intel CPUs. Once we’ve gotten through the i3s and other i5s, we should better know which make the most sense at each price-point, and which become embattled with AMD at each price. For now, the i5-8400 performs reasonably and exhibits +/-5-7% performance swings from memory speed differences (3200MHz vs. 2666MHz) in most games, but we’d advise waiting. It doesn’t make any sense at all to buy a Z-series motherboard – the only 300-series boards presently out – with a locked i5 CPU; at least, not unless you’re planning to buy a K-SKU part later.
So much for Steve trying to do a 'sales piece' for Z370! LOL! If he was, he is doing a terrible job.
 

krumme

Diamond Member
Oct 9, 2009
5,952
1,584
136
Seriously, did you even bother to read the article that goes along with the video? Obviously not, otherwise you wouldn't be making such accusations:



So much for Steve trying to do a 'sales piece' for Z370! LOL! If he was, he is doing a terrible job.
Yes and I am saying this performance increase is probably there because of latency/speed factor. The test need to separate what the cause is instead of saying it's because of bandwith difference. 5 to 7% perf difference is something. It's actually interesting how those 6c cpu hides latency.
It's not always I also read the entire article and I am sure in am not the only one. I tried to find the latency numbers. By far the most interesting. Yet he chooses not to disclose it but hides it away. Wtf can you use the results for then?
 

epsilon84

Golden Member
Aug 29, 2010
1,142
927
136
Yes and I am saying this performance increase is probably there because of latency/speed factor. The test need to separate what the cause is instead of saying it's because of bandwith difference. 5 to 7% perf difference is something. It's actually interesting how those 6c cpu hides latency.
It's not always I also read the entire article and I am sure in am not the only one. I tried to find the latency numbers. By far the most interesting. Yet he chooses not to disclose it but hides it away. Wtf can you use the results for then?
As I said earlier, just leave a comment under the relevant video, you might get all the info you need re: memory latency right there

FWIW, Intel platforms have been sensitive to bandwith when it comes to gaming for a long time. Tight timings do help of course but overall memory bandwith seems to help even more.
 

Bouowmx

Golden Member
Nov 13, 2016
1,138
550
146
FWIW, Intel platforms have been sensitive to bandwith when it comes to gaming for a long time.
I am under the impression that once there is enough bandwidth, any more does not help. Better to reduce latency. 2-channel 2400 or 2666 MT/s appears to be the point of enough bandwidth, for i7-8700K.
 
  • Like
Reactions: krumme

Glo.

Diamond Member
Apr 25, 2015
5,245
3,813
136
"as of now, we are not sure whether B/H boards will support only the native memory speed of 2666MHz or higher multipliers. Some board vendors have suggested a possibility of unlocked memory multipliers of 32/36x, but haven’t confirmed, while other sources have suggested a maximum speed of 2666MHz. Because we cannot reasonably confirm either, we decided to just test both, then let the chips fall where they may in 1Q18."

Source - Gamers Nexus 8400 2666/3200mhz review
This suggests one thing. 2666 MHz on B and H MoBo and Core i3-8100, and Core i5-8400 will be possible PnP.

If wil be possible 3200 MHz on any of ASRock MoBos with mITX - im building new computer. Finally.
 
  • Like
Reactions: Crono

VirtualLarry

No Lifer
Aug 25, 2001
54,961
9,023
126
This suggests one thing. 2666 MHz on B and H MoBo and Core i3-8100, and Core i5-8400 will be possible PnP.

If wil be possible 3200 MHz on any of ASRock MoBos with mITX - im building new computer. Finally.
I've got 2800 RAM on my ASRock Z370 ITX/ac with i3-8100, using XMP.
 
  • Like
Reactions: krumme

ASK THE COMMUNITY