AMD Ryzen's First Game Optimization: 'Ashes Of The Singularity: Escalation,' Tested

Status
Not open for further replies.

Jakko_

Prominent
Mar 28, 2017
7
0
510
"The update is our first example of a game optimized for AMD's Ryzen CPUs, which should improve their performance."

Not true, dota 2 was the first. I hope you guys will consider adding this game to your Ryzen 5 review, it IS the most popular online game in the world.
 

ClusT3R

Distinguished
Dec 29, 2009
16
0
18,510
Like I always said there is a beast sleeping in those CPU how will perform overclocked ? specially the 1700 at 4Ghz OC put better result in gaming than the 1800x
 

falchard

Distinguished
Jun 13, 2008
2,360
0
19,790
Ryzen sees massive gains going from 2133mhz to 3200mhz. Bios still not at a point where this is stable. Also the temperature reading loves to fluctuate. Either really accurate, or its the self overclocking.
 

dasick

Honorable
Dec 26, 2012
5
0
10,510
"The 1800X enjoys a 7% performance boost from the update and jumps up several positions."

How did you come up with 7%?
My English is not to good but I know the difference between "al least 7%" and "7%".

1800X CPU Test 1920x1080
New: 45.5
Old: 35.1

(45.5-35.1)/(35.1/100)=15.43%

This 15.43 is about the same percentage other sites have come up with (15-16%), only Toms calculated half the acceleration.
I usually don't comment this kind of "mistakes" and I'm really not a conspiracy guy, but reading Rayzen articles on Toms I can't help impression that you guys are not objective in AMD vs Intel "race".

Of course this is just my opinion.
 

dasick

Honorable
Dec 26, 2012
5
0
10,510
PS.
I don'd know how to edit post here.
But its early here and I haven't had my morning coffee yet : )
Calculation that I left showed about 30%, and 15% I calculated is for "Total average CPU Test 1920x1080" (40.47-35.08)/(35.08/100)= ~15%

Sorry, my mistake.
 

t1gran

Prominent
Mar 6, 2017
10
0
510
What is the difference between CPU Test (Ryzen 7 1800X New - 40.5 fps) and GPU Test (60.3)? Aren't they both average fps of Ryzen 7 1800X coupled with GTX 1080 in Ashes of the Singularity: Escalation?
 

bwcbwc

Distinguished
Apr 28, 2010
41
0
18,530
Seeing these results, I wouldn't call it an AMD-specific optimization. AMD gains the most, but the Intel processors also see small improvements in FPS, and the 7600k even improves its frame-time variance. In an industry that loves to squeeze out every resource in optimizations, this might encourage developers to take a closer look.
 

hannibal

Distinguished
All "real" gamers play with 1080p with 167383839 frame per second using the easiest graphic settings They can find. But normal user will use all the bells and whisles and use the highest possible settings, so this has some validy to some heavy gamers.

The important thing is that if programmers take care, Ryzen can be a hell a lot good Gaming CPU also. Lets hope that programmers consider this, because it will mean cheaper cpus to players and so more competition and more posibilities to game programing when x core CPU Are common from all cpu makers.
 

ohim

Distinguished
Feb 10, 2009
1,195
0
19,360
This proves two things :

1. There is much to be gained when devs optimize code for Ryzen
2. Don`t trust that much a benchmark when doing a purchase, a badly optimized benchmark can favour a component more than other...
 

Rogue Leader

It's a trap!
Moderator


Not as true anymore. Ram with Samsung B-Die has been stable for a while. ASUS just released a BIOS update for the Crosshair VI that totally updated the memory compatibility. I am running my Corsair Vengeance LED DDR4-3200 at full speed with no issues on there now and its SK Hynix based.
 

BonScott

Prominent
Mar 30, 2017
44
0
530
Great comparison. I understand the system needs to be the same to compare SW optimization, but one thing that people have learned about Ryzen is that when 1080p gaming w/ the fastest graphics card available (GTX1080 or 1080Ti) having a system that can support 3200MHz DDR4 clocks, or at least 2933MHz, is what corrects for some of the large FPS deficiencies seen with some games.

Beyond that, 8-core Ryzen is pretty close in FPS, despite lower clock and optimizations still coming ... and 6 and 4 core Ryzen coming. Really those are better matched comparisons to 7700K.

Beyond 1080p gaming with a GTX 1080 (which is becoming less of an issue), the Ryzen wins on performance, power, and price.
 

BonScott

Prominent
Mar 30, 2017
44
0
530
these tom hardware benches are at 2666MHz DDR4 for Ryzen, so not as bad as 2133MHz, but 2933Mhz better, and 3200MHz better yet, and seemed to be the threshold in some games to address 1080p gaming cases where Ryzen's FPS gap was unexpectedly larger compared to other general benchmarks.
 

Josh_194

Prominent
Mar 30, 2017
4
0
510
Please stop testing CPUs by benchmarking video games. Bottlenecking the GPU doesn't eliminate enough variables to analyze how well the CPU performs. This is bad test engineering that showed performance issues stemming from software/hardware interaction. The test community falsely claimed Ryzen was "bad" for gaming, and gave millions of consumers bad data points.
 

TJ Hooker

Titan
Ambassador

If you're saying they should test non-gaming application performance, there's plenty of that in the actual Ryzen 7 CPU reviews.
If you're saying benchmarking video games is not a good way to measure video game performance... what? Or if there's something specific about the way that they're benchmarking that you don't like, you'll have to elaborate.

Most people didn't say that Ryzen was bad for gaming. They said that Ryzen 7 CPUs are typically beat by Intel i5/i7 quads (which are cheaper), meaning that Ryzen 7 CPUs offer worse performance per dollar if all you're doing is gaming. Which is exactly what is said about Intel's LGA 2011 CPUs for the most part. Once Ryzen 5 comes out, I imagine we'll see more recommendations for Ryzen CPUs in gaming-centric builds.
 

80-watt Hamster

Honorable
Oct 9, 2014
238
18
10,715


There has been plenty of non-game application testing performed, including here at Tom's. Ryzen was very competitive in many of those tests, while consistently falling behind Intel chips in games. That sort of thing makes people wonder why, and justifies further testing. The fact that software tweaks have shaken up the results tables suggests that the investigation was worth undertaking.

Now, I agree that there is a marked tendency in enthusiast communities to label anything that's "not the best" as "bad"; the gaming world seems particularly susceptible to this. Partisanship aside, though, I don't see what's dishonest about coming to the conclusion that Intel's high-end processors were, in a given set of circumstances at a particular point in time, superior to AMD's when CPU performance was a limiting factor. While you're absolutely correct that Ryzen is not a bad processor for gaming, it's also true that a 7700K is still the better choice (so far) in performance/dollar for that specific purpose, if not by much.
 

Sam Bittermann

Honorable
Aug 15, 2013
37
0
10,530
these tom hardware benches are at 2666MHz DDR4 for Ryzen, so not as bad as 2133MHz, but 2933Mhz better, and 3200MHz better yet, and seemed to be the threshold in some games to address 1080p gaming cases where Ryzen's FPS gap was unexpectedly larger compared to other general benchmarks.

I'd like to know the answer to why Tom's did not use the higher bandwidth memory as Ryzen performs better with faster memory? Is that what they used previously to make it apples to apples?
 

SmnOOthStar

Prominent
Mar 30, 2017
1
0
510
I'm not expecting AMD Ryzen to become the best choice for gamers. It's said the architecture of new Ryzens has been developped for various of tasks without focusing on gaming. I appreciate its versality and still I find Ryzens interesting choice for the money. Still if it's just about how developer will work out optimalization we could get even better benchmarks in time.
 

InvalidError

Titan
Moderator

Because this is a CPU benchmark and to see how quickly the CPU can setup the next frame, the GPU has to be as little of a bottleneck as possible. The ideal GPU for this purpose would be a NULL driver to simulate an infinitely powerful GPU. Since no such driver is available, the next best thing is lower the resolution to where it isn't likely to be a bottleneck.

Benchmarking a CPU is pointless when frame rates are limited by the GPU.
 

hannibal

Distinguished
Faster memory is normally Also more expensive, so using officially supported ratings in the memory Sounds very plausible instead of more expensive and unofficially supported, but faster memory timings.

What would be interesting is a test with different memory speeds, but that can wait Until the mother board bioses gets some more upgrades! It would be usefull to know where the sweet spot for the memory is for Ryzen!
 
Status
Not open for further replies.