Coffeelake thread, benchmarks, reviews, input, everything.

Page 47 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
O

ozzy702

Golden Member
Nov 1, 2011
1,151
530
136
DaveSimmons said:
Interesting, so you'll get maybe 1 more FPS going from 2666 up to 4000. Faster than 2666 seems like a waste of money for gaming on Coffee Lake CPUs.
Click to expand...

The problem with that test is it's using a 1080, and only showing average FPS, not frame times or lows. They really should have used a 1080ti at max OC. I still doubt there would have been huge differences but I'm sure there would be cases where the system benefited from faster memory, especially in frame times.

I have two profiles on my 8700k with a 1080ti. A 24/7 profile undervolted and underclocked @ 3.9ghz and 2133mhz memory and a 4.8ghz 3400mhz. I started up witcher 3 the other night forgetting I was using the slower profile and started noticing micro-stutter and a general lack of smoothness despite pulling over 100fps and using gsync. It dawned on me that I wasn't using my gaming profile and so I rebooted, fired up the game and boom, everything was silky smooth again. Yes, I realize that 2133mhz is dog slow, but even at 2666 I bet the same micro stutter would have been present.
 
Last edited:
  • Like
Reactions: hnizdo
Arachnotronic

Arachnotronic

Lifer
Mar 10, 2006
11,715
2,012
126
ozzy702 said:
None of that makes any sense. If quad channel memory mattered in gaming then the 6900k would be besting the 8700k but it doesn't come close in anything but possibly Ashes. 8700k has significantly lower latency than 6900k, higher IPC and faster clock speeds, the delta would only widen with two additional cores.

Power usage is power usage. A CPU using 200w with TIM will "dump heat into the case" at roughly the same rate as a 200w CPU with solder. Yes, if you're using a large water cooler it may be able to dump slightly more heat out of the case with a soldered CPU but the difference is negligible. The fact of the matter is that at max OC the 8900k would use less power than the 6900k (~215W @ 4.3ghz) and be significantly faster in every single way.
Click to expand...

Memory latency is more important to gaming performance than bandwidth.
 
  • Like
Reactions: Edrick, ZGR, CHADBOGA and 1 other person
VirtualLarry

VirtualLarry

No Lifer
Aug 25, 2001
56,349
10,049
126
VRM heatsink issues? I have to ask, WHY? I mean, CPUs back in the Core2Quad days, regularly took 100-115W TDP, in real life, and the VRMs managed fine, but back then, the VRM heatsinks were more than just decorative, on high-end boards, they had heatpipes between the chipset and VRM heatsinks.

Now, CPUs take LESS power (or maybe not, when overclocked), and have beefier VRMs, and yet, they overheat MORE? What gives? Why have mobos gone backwards? Cost-cutting?
 
  • Like
Reactions: pcp7 and Drazick
dullard

dullard

Elite Member
May 21, 2001
25,069
3,420
126
VirtualLarry said:
VRM heatsink issues? I have to ask, WHY? I mean, CPUs back in the Core2Quad days, regularly took 100-115W TDP, in real life, and the VRMs managed fine, but back then, the VRM heatsinks were more than just decorative, on high-end boards, they had heatpipes between the chipset and VRM heatsinks.

Now, CPUs take LESS power (or maybe not, when overclocked), and have beefier VRMs, and yet, they overheat MORE? What gives? Why have mobos gone backwards? Cost-cutting?
Click to expand...
It doesn't quite work like that. A 105 W Q6600 at about 1.5 V is 70 A. A 95 W 8700K at about 1.25 V is 76 A. Even though the power is slightly lower, the current is the same or even a bit higher. I believe with voltage regulators what matters is the current that they have to produce.

But your point is correct that these may have been skimpy.
 
  • Like
Reactions: pcp7
DrMrLordX

DrMrLordX

Lifer
Apr 27, 2000
21,637
10,855
136
Gigabyte had problems with VRMs on some of the X370 and B350 boards as well. The Gaming-3 has the 4+3 layout for Bristol Ridge for crying out loud. I don't know what's been wrong with them lately.
 
VirtualLarry

VirtualLarry

No Lifer
Aug 25, 2001
56,349
10,049
126
dullard said:
It doesn't quite work like that. A 105 W Q6600 at about 1.5 V is 70 A. A 95 W 8700K at about 1.25 V is 76 A. Even though the power is slightly lower, the current is the same or even a bit higher. I believe with voltage regulators what matters is the current that they have to produce.
Click to expand...
You are correct, sir. I didn't quite think it through.

Modern CPUs operate at a lower voltage, so at the same power levels (95W TDP), they are drawing MORE current, and temps have to do with current per cross-sectional area, I believe, as far as wire-heating goes, so there's that.
 
  • Like
Reactions: Drazick
LTC8K6

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
Asterox said:
Well, this gaming comparison is "not very good" for i7 8700K or ultimate gaming King.:D

http://www.hardwarecanucks.com/foru...7-2600k-vs-i7-8700k-upgrading-worthwhile.html
Click to expand...
Most games not taking advantage of more than 4 cores is not really news, though.

It doesn't bode well for upgrading to any CPU with more than 4 cores / 8 threads if you are gaming centric.

You certainly don't need a 16 thread CPU for gaming.

Apart from gaming, more modern cores, and more of them, do seem to now be a good advantage in many areas, and are cause for an upgrade of an older CPU.
 
B

beginner99

Diamond Member
Jun 2, 2009
5,210
1,580
136
ZGR said:
Cpu test with Ultra settings... WTF. GPU limited in nearly every benchmark for games.

Why even bother? Ultra settings is lazy. :(
Click to expand...

that + the fact that BF1 single player <> BF1 64-player

Or completely different end of the scale is Starcraft 2 where the higher IPC and especially higher clocks come into play.

Sure, if all you do is play Single-player at 1080p 60hz, a 8700k is overkill. But no one very claimed otherwise.
 
  • Like
Reactions: ZGR
CatMerc

CatMerc

Golden Member
Jul 16, 2016
1,114
1,149
136
Huh

My AX370 Gaming K7 VRM's barely heat up. They stay in the 60 to 70 area under heavy load.
 
D

Dribble

Platinum Member
Aug 9, 2005
2,076
611
136
TahoeDust said:
Watch their follow up when they ran their tests with a 1080ti...

Click to expand...
I think he misses the point. Do you think when the 2600K came out you needed that cpu? I bet if he'd done the same test then and been gpu limited with his geforce 6800 or whatever then an o/c Conroe Q6600 was probably giving close to the same frames. Yet 6.5 years on that 2600K is still good, where as the Q6600 is way past it.

You buy a high end cpu looking to the future - it's hard to upgrade a cpu so you don't want to do it often (any meaningful upgrade from a high end cpu needs a new mb and memory which is basically a whole pc rebuild). Hence then we bought 2600K's, now we buy 8700K's. Just because it's not really stretching it's legs in today's games doesn't mean it won't in the future, and what we want to know is how much performance is there so we can tell how future proof it is. Just like then we wanted to know how much faster a 2600K was over a Q6600 when you weren't gpu limited so we'd know the 2600K would still be kicking it 6.5 years later, and not dog slow like the Q6600.
 
  • Like
Reactions: Campy, Zucker2k, IEC and 1 other person
DrMrLordX

DrMrLordX

Lifer
Apr 27, 2000
21,637
10,855
136
CatMerc said:
Huh

My AX370 Gaming K7 VRM's barely heat up. They stay in the 60 to 70 area under heavy load.
Click to expand...

Some folks (like chew) have reported problems with the K5's power delivery. Blowing out boards and stuff. The K7 might have better sinks, who knows?

My Taichi has VRMs in maybe the 50C range while stress testing, which is awesome.
 
H

hnizdo

Member
Aug 11, 2017
33
16
41
Asterox said:
Well, this gaming comparison is "not very good" for i7 8700K or ultimate gaming King.:D

http://www.hardwarecanucks.com/foru...7-2600k-vs-i7-8700k-upgrading-worthwhile.html
Click to expand...

Results depends on game and scene selection. There were tests, where GTA-V had 50% fps in ivy bridge to skylake (!) comparison on GTX 970 (!). It seems like clickbait to me rather than serious test.
I have completely different gameplay in Arma3/GTX970 after upgrade i7-3770K@4.1 to 8700K, despite average fps didnt rise several times. But it feels like it does.
 
S

slashy16

Member
Mar 24, 2017
151
59
71
Asterox said:
Well, this gaming comparison is "not very good" for i7 8700K or ultimate gaming King.:D

http://www.hardwarecanucks.com/foru...7-2600k-vs-i7-8700k-upgrading-worthwhile.html
Click to expand...

There are plenty of actual comparisons showing real side by side performance between all of the i7 generations and there is large performance gaps between the 2600k and the 7700k depending on the title. I am one of those people who have a 1080ti and also run 1080p because I have three 27" monitors that only run 1080p and I need to replace them. I got a 1080ti because I play games like world of warships that can have up to 24 players displayed on the screen and when you add in smoke effects the framerate absolutely tanks. Anyone running a 2500/2600 should seriously consider upgrading to a 8600/8700 or even Ryzen if they are on a budget. Low frame dips and overall system smoothness are much improved. Just doing simple things like running two monitors so you can watch twitch and play PUBG at the same time are worlds better on newer processors. My brother in law bought my i7 3770k and upgraded from a 2500k clocked at 4.2ghz to my i7 3770k@4.6ghz and the overall system experience was much improved.
 
  • Like
Reactions: hnizdo
ehume

ehume

Golden Member
Nov 6, 2009
1,511
73
91
CatMerc said:
Huh

My AX370 Gaming K7 VRM's barely heat up. They stay in the 60 to 70 area under heavy load.
Click to expand...
Is that with the little fan running or with it shut down? I am looking for a quiet board so I can compare heatsinks.
 
You must log in or register to reply here.

TRENDING THREADS

COMPANY
RESOURCES
FOLLOW
Top Bottom