It's comparing draw call performance, ya nunce. Poor draw call performance + major CCX penalties + ~Haswell number crunching performance = Not so good gaming chip.
But 'ey, if you can find me ten guys to run my Fallout 4 benchmark, we'd be golden.
Now show me the correlation with that and actual games run at realistic settings. You can't, because that information is about as useful as cinebench for making predictions.
Yet, that isn't what the real world data says. Which means your prediction model doesn't work. Isn't science great?It's comparing draw call performance, ya nunce. Poor draw call performance + major CCX penalties + ~Haswell number crunching performance = Not so good gaming chip.
But 'ey, if you can find me ten guys to run my Fallout 4 benchmark, we'd be golden.
Yet, that isn't what the real world data says. Which means your prediction model doesn't work. Isn't science great?
Why? That sorry excuse of a modern game isn't relevant... Heck, even when it was relevant, people still avoided benchmarking it, because it runs so bad.
Since we are benchmarking old games anyway, we should benchmark world of warcraft and age of empires too.
You've seemed to have flipped the scientific method on it's head. You already have your conclusion set, and you are looking for data to support it.
If you're basing gaming performance on that graph, then all PC gamers should be buying $75 dual core pentium processors.
Ryzen's not all that close to Haswell. In single threaded workloads, sure. But in gaming, the cross-CCX draw call performance is abysmal; it's at Core 2 levels. On a single CCX, with 3200Mhz DDR4, Ryzen performs a tad better than Sandybridge at draw calls.
It would take even faster RAM for AMD to catch up to Haswell in intensive scenes, and if it uses more than four cores...Ew.
I'm not sure I understand. I have a Xeon Haswell 1231 running all cores at 3.8. It's about 20-30% slower than my Ryzen 1700 at 3.7. In multithreaded apps it's a total wash for the Ryzen.
Intel has good chips, but no need to go hyperbole.
And of course you are ignoring the new benchmarks@3600mhz where Ryzen shines everywhere and beats all but in one game.We're talking about games bruv. Did you check the cross-CCX scores? It's marginally faster than Phenom II when coupled with 2133Mhz RAM, with the driver and game thread on different CCX units. Alleviated a bit by having 3200Mhz RAM, but that's only achieving parity with Core 2.
And of course you are ignoring the new benchmarks@3600mhz where Ryzen shines everywhere and beats all but in one game.
@MajinCry You run it. And you tell me whats wrong with all the OTHER games it does well at.(like all of these) You are the one maintaining that it does poorly on games, so you prove your point.
Majin, I think your draw call benchmark is legitimate but I also don't know how applicable it is to how modern games are coded (Fallout 4 does not fall into this category, imo) and it also doesn't take into account that most games don't fall into such a heavily weighted draw call bottleneck. Have you thought about throwing a more modern open world game into the mix? Mafia 3 is an open world game that Ryzen seems to do really well in. It would be interesting to see it benched in a heavy draw call situation in that game or another similar title.
Might want to check the date of that draw call benchmark, sonny.
And surely a game that runs bad, due to it having so much going on at once (a good few thousand actors being processed at any one time, draw calls in the high thousands, immense draw distances), is the perfect candidate to show the muscle of an architecture?
You'll find that World of Warcraft is also a prime candidate for testing. Pretty much every MMO that supports widescale PvP is, actually. Shows how much strength is in the CPU, as those games are extremely heavy on the renderer.
It's curious that you find a game from 2015 to be "old", by and by. I'd consider Neverwinter Nights or Morrowind to be old.
You can say w/e you want and in any condescending tone that you want, but you're still wrong. Your model doesn't work.
It might have been made in 2015, but the coding is much older. They practically used an antique engine.
So, you admit your model doesn't work on modern titles.So, to clarify, you're absolutely fine with poor performance in pre-existing games (the whole point of x86 computers being legacy support), as long as bleeding-edge games perform acceptably?
We're talking about games bruv. Did you check the cross-CCX scores? It's marginally faster than Phenom II when coupled with 2133Mhz RAM, with the driver and game thread on different CCX units. Alleviated a bit by having 3200Mhz RAM, but that's only achieving parity with Core 2.
Poor performance ? where are these "bad benchmark" numbers ? All you cry is about draw calls. I could care less about something that does not prove a point. What works badly ? Where is your proof ? Where are your benchmarks. This is trolling, and you keep it up, and you are going to get infracted.So, to clarify, you're absolutely fine with poor performance in pre-existing games (the whole point of x86 computers being legacy support), as long as bleeding-edge games perform acceptably?
Poor performance ? where are these "bad benchmark" numbers ? All you cry is about draw calls. I could care less about something that does not prove a point. What works badly ? Where is your proof ? Where are your benchmarks. This is trolling, and you keep it up, and you are going to get infracted.
How much relevance does a DX9 draw call test and a DX9 game have nowadays for performance looking forward? Maybe if you gathered data from 3dMark API overhead test for DX12+Vulkan it might be more relevant?I've already provided the benchmark, with a graph of the results posted right here in this thread. Just because you are unwilling to consider Ryzen's decade old draw call performance, doesn't magically make Ryzen perform better at draw calls.
+1. Exactly what I was thinking.How much relevance does a DX9 draw call test and a DX9 game have nowadays for performance looking forward? Maybe if you gathered data from 3dMark API overhead test for DX12+Vulkan it might be more relevant?
This forum seems to skew heavily toward gamers. Those of us who do actual serious work and/or whose idea of a game is "what does Gentoo think of this bad girl?!" are in heaven with Ryzen. Honestly I don't feel the chip got a fair shake in most reviews.
How much relevance does a DX9 draw call test and a DX9 game have nowadays for performance looking forward? Maybe if you gathered data from 3dMark API overhead test for DX12+Vulkan it might be more relevant?