Ryzen-A Fail for Gamers?

Page 10 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

unseenmorbidity

Golden Member
Nov 27, 2016
1,395
967
96
One thing is for sure.
One is not going to have games that work well on 8+ cores until you have a decent user base with 8+ cores.

About the 1080p.
If you go to the video card forums and ask about a card for 1080p most people will point you into the direction of a GTX 1060 or RX 480.

Correct,

Except, games are made for consoles. That is the platform the game developers care about, and that platform has 8C/8T. The PC market is an afterthought, and mostly relegated to mediocre ports.

Bethesda isn't going to care that you have to turn shadows down on your i5. Rockstar is not going to care that you need to lower the population density in GTAVI.
 
Last edited:

HiroThreading

Member
Apr 25, 2016
173
29
91
It's a joke if you are buying a cpu just to play today's games. If you are buying however to play games over the next few years, this is a good way to predict how each cpu will do when games get more demanding and GPUs get stronger. The weaker cpus in this test will falter first when that happens.

Note, I am saying this as someone who bought an 1800x, not as someone going 7700k.

Yeah I agree with that statement. But at worst Zen performs like a 6-8 core Broadwell would against a 4 core Kaby Lake, and at its best it's top of the pack with its 16 threads -- something that will benefit it (as well as Intel's HEDT lineup) going into the future. Broadwell level performance is bloody fast. It's not like Bulldozer or Piledriver at all -- there really isn't any major compromise here with Zen, so to speak.

It's like back in the day when when Intel cut prices on the Core 2 Quad Q6600 Kentsfield. At worst, it would perform like a Conroe, but at its best it ran circles against equivalently priced Wolfdale chips (C2D E8000 series dual cores). Conroe performance was nothing to sneeze at, but the future potential upside of Kentsfield made it the easy recommendation over the then-faster Wolfdale chips. Looking back with the benefit of hindsight, I think the people who chose Kentsfield over Wolfdale were vindicated in their decision. Those Q6600s lasted until about Sandy Bridge/Ivy Bridge.
 

Headfoot

Diamond Member
Feb 28, 2008
4,444
641
126
For gaming I want to see a best-config vs best-config shootout with modern games.

7700k @ 5ghz with DDR4 3600 vs. 1700 @ max oc (4.0-1 ghz?) with DDR4 3600, HT off. The overclocker should spend the time seeing if there are things on Ryzen that pull up performance but aren't obvious - Bus overclocking to get more RAM speed, L3 overclocking if possible, uncore overclocking if possible. Both on water cooling.

I just learned recently I'm leaving performance on the table with my 5820k because I haven't overclocked the L3. I think we need to explore the other tweaks available on Zen besides just multiplier overclock. I only reached my own OC at such low voltage because I fooled around with bus overclocking and was able to hit the same (slightly higher) OC with less voltage via that route. Also I have a nice chip.

As of Skylake RAM is back in the game, some reviewers have picked up on that but not all.

I want to see the true hardware enthusiasts fully tweaked shootout.
 

EXCellR8

Diamond Member
Sep 1, 2010
3,979
839
136
More like "PC... a fail for gamers"

seriously... no half life 3, very few cryengine games, lousy ports everywhere, and steam is an absolute dumping ground for paid, half-baked indie projects that almost never fulfill promises.

depressing.
 

french toast

Senior member
Feb 22, 2017
988
825
136
It's a joke if you are buying a cpu just to play today's games. If you are buying however to play games over the next few years, this is a good way to predict how each cpu will do when games get more demanding and GPUs get stronger. The weaker cpus in this test will falter first when that happens.

Note, I am saying this as someone who bought an 1800x, not as someone going 7700k.
Debunked!!
 

guachi

Senior member
Nov 16, 2010
761
415
136
Do you have a video of that tester actually testing a 7700K in that section of the game with a 1070 GPU at 1920x1080 ultra settings?

The video you linked is just of the R7 1700.
 

Head1985

Golden Member
Jul 8, 2014
1,863
685
136
R7 1700 is likely running at 3.2ghz if not overclocked, its also gpu bound, plus known bugs for new uarch, what do you expect?
well better performance:D
Witcher3 can use 12-16threads.Btw its not GPU bound..GPU usage dips into 80-90%
 

Head1985

Golden Member
Jul 8, 2014
1,863
685
136
here is my run in witcher3 with GTX1070 oc same time and details.You can see 99% GPU usage all time
 

french toast

Senior member
Feb 22, 2017
988
825
136
well better performance:D
Witcher3 can use 12-16threads.Btw its not GPU bound..GPU usage dips into 80-90%
GPU load regularly gets pinned at 98% for long periods, just look at the cpu load on the threads, hardly breaking a sweat!
Besides fps is much better than barely hitting 60fps, hitting mid 80s at various points.
Its gpu bound no question, the game and windows need patching for ryzens new uarch.
 

french toast

Senior member
Feb 22, 2017
988
825
136
His 1070 is 2101/9100
Mine is 2114/9500
You stop trolling.Clearly i dont have any GPU bottleneck he have and i have way more fps.
You are clearly trolling.
Your "test" at unknown clocks/ram-speed-timings/cooling vs a completely unknown setup and bios.

Your "test" averages about 84/85fps with dips into low 70s.
Ryzen "test" averages about 72/73fps with dips into the 60s.

Thats hardly twice the speed your were trying to make out now was it? Where is this 120fps you were taking about? Rubbish.
Considering ryzen is clocked so low, is a 65w chip, likely running crap memory on an unoptimised bios, with broken windows driver and no game patch to get proper load balancing....Ryzen actually does better than i imagined.

Stop trolling.
 
Last edited:

Head1985

Golden Member
Jul 8, 2014
1,863
685
136
Last edited:

unseenmorbidity

Golden Member
Nov 27, 2016
1,395
967
96
120fps with faster GPU omg.I am 100% time GPu bottleneck with GTX1070 he is not he have cpu bottleneck.
you didnt watch digital foundry videos?
https://youtu.be/tbGT-u4i3EY?t=714

ryzen have 85% GPU usage in that market with 60-65fps=He is cpu bottleneck.
Watch his GPU usage
https://youtu.be/O-iEpSRey1s?t=129

Here is mine video 99% GPU usage
https://youtu.be/5bNpBTS24NI?t=20

Anyway lets wait for digital foundy review.You will see.
You cannot do a random comparative benchmark without know every setting that is used on the other system... You are talking out of your arse.
 
  • Like
Reactions: french toast

Head1985

Golden Member
Jul 8, 2014
1,863
685
136
60 somethin vs 80ish. Meh, screw it. No biggie really. Difference would be much less vs 6900k.
With GTX1070..With TITANXP/1080TI its 120-130fps vs 60-65fps.He have there cpu bottleneck.Better GPU wont help ryzen.
 

french toast

Senior member
Feb 22, 2017
988
825
136
With GTX1070..With TITANXP/1080TI its 120-130fps vs 60-65fps.He have there cpu bottleneck.Better GPU wont help ryzen.
A cow is better than a sheep because its can eat more grass, the sheep has a chewing bottleneck.
Makes as much sense as your last ten posts.
 
  • Like
Reactions: guachi

Head1985

Golden Member
Jul 8, 2014
1,863
685
136
Just read somewhere how CPU/GPU bottleneck works then come back.If he is already on CPU bottleneck with 1070 then faster GPU wont help.Only OC cpu will help.
7700k without GPU bottleneck is able push 120-130fps in that part.
https://youtu.be/tbGT-u4i3EY?t=720
Why is here so many people whos dont know anything how CPU/GPU bottleneck works?