[TH] Ashes of the Singularity - Ryzen Update (15% Boost!)

[nathanddrews]

According to AMD, games get a 31% boost at 1080p High for All Batches.

Of course Tom's benched 1080p Crazy instead, which saw a jump from 35fps to 40fps (~15%), which now places it in the lead over the 7700K (which also improved slightly).

Despite Tom's title, Dota 2 was the first known game to get patched for Ryzen and showed a ~19% improvement.

 

[nathanddrews]

I didn't realize that PCPer already benched this two days ago. They actually hit the 31% claim!

 

[Atari2600]

I think its up to 31% based on settings.

Other settings shown an improvement of ~20%.

 

[nathanddrews]

I think its up to 31% based on settings.

According to AMD, games get a 31% boost at 1080p High for All Batches.

Indeed.

 

[Malogeek]

What's with the 8 cores terrible performance?

 

[nathanddrews]

What's with the 8 cores terrible performance?

Link to the full article and maybe they say?

 

[Malogeek]

Link to the full article and maybe they say?

http://www.pcgameshardware.de/Ryzen...ecials/AMD-AotS-Patch-Test-Benchmark-1224503/

From the translations they're not really sure. Citing issues with save games compatibility across major versions being a problem it's difficult for them to narrow down apparently. I'll have to find more sources as other places re-review with the patch and have 8 core parts to compare with.

 

[nathanddrews]

Definitely strange about the savegame issue, but hasn't that always been problematic before this patch as well?

The integrated CPU and GPU benchmarks show significant improvements for Ryzen and mild improvements for the 6900K:
GPU: Ryzen went from being 12% behind the 7700K to being 9% ahead. From 30% behind the 6900K to being 8% behind.
CPU: Ryzen went from being 9% ahead of the 7700K to being 23% ahead. From 32% behind the 6900K to being 21% behind.

 

[unseenmorbidity]

What's with the 8 cores terrible performance?

i heard they only optimized the benchmark, and not the actual game.

 

[Malogeek]

i heard they only optimized the benchmark, and not the actual game.

 

[Hitman928]

What's with the 8 cores terrible performance?

This is exactly why I have a hard time trusting review sites anymore. The 1800x results initially should have had a huge red flag that something was wrong with their test setup but instead of investigating and running some 'sanity checks' to verify the results they just publish anyway and then later realize something is wrong when the 6900k is even worse. If they hadn't bothered to test the 6900k eventually, people would have (and did) think that the game just runs terribly on ryzen. I've started to treat reviews kind of like peer reviewed journals. I don't trust any results until multiple other sites consistently get the same results. I realize there's a need to be first to publish in today's world but it shouldn't come at the expense of your integrity which has happened to far too many sites. /rant

 

[Dresdenboy]

i heard they only optimized the benchmark, and not the actual game.

OK, let's wait for feedback from actual gamers.

 

[AMDisTheBEST]

Total war series need serious patch. Amd chips run like absolute $@#! on them.

 

[IEC]

What's with the 8 cores terrible performance?

AotS automatically reduces details in-game (not in benchmark) on systems with fewer than 6 physical cores, no matter what settings you choose.

Source:
http://www.pcgameshardware.de/Ryzen...ecials/AMD-AotS-Patch-Test-Benchmark-1224503/

Short summary of important points for those of you who don't read German:
-Also affected i7-6900K framerates, which made them suspect something was off
-The results they saw were weird, so they retested, including SMT on/off. No difference.
-Tested AVX2 by manually downclocking AVX2 on the Intel HEDT chip and on the Kaby Lake 7700K. No difference.
-Finally discovered that fewer than six physical cores (disclaimer, they did not test 5 physical cores) results in the game removing entire elements such as enemy defensive fire particle effects from the game.
-Tested on a simulated 4 core Ryzen "1500X" and discovered same performance uplift. Performance versus 7700K within margin of error.
-Conclusion: Benchmarks are only comparable quad to quad and octa to octa.

 

[Malogeek]

AotS automatically reduces details in-game (not in benchmark) on systems with fewer than 6 physical cores, no matter what settings you choose.

Yep just heard that. So basically it's impossible to use anything in-game for benchmarking.