Can you bench it with this thread demo?

Could you benchmark with mastercomfig low? Wondering how it'll work on these CPUs.

I'll do both tomorrow
i need someone to send me the old benchmark demo

Someone has reuploaded it in previous posts.
https://www102.zippyshare.com/v/fvfg8CAP/file.html

Not sure what you mean. My 8350K is pounding all these Zen 2 chips in TF2 performance. Unless you mean you regret not getting a much better overall chip that still has decent TF2 performance.

really?
i get around 200-210 fps in the mastercoms benchmark (with mastercoms medium-low) at 4ghz (i oc'd to 4.5 for a bit and it wasn't that much of a difference), nowhere near as close as the 300+ fps people are posting here

RAM matters a lot too for TF2. I have B-die at 4133 which allows >290 FPS in benchmark1.dem. You should OC to at least 4800 on your 8350K, which can be achieved with a fair voltage. 4900 and above start to require excessive vcore.

Which one do you mean?
Moist running a stock 3600 with 3200 RAM, benching a completely different config and demo and getting 360 fps? How is that getting pounded by the 8350K?
Or do you mean mousiope's stock 3600X again with 3200 RAM, no config and highest settings getting 190 fps? I guess 300 fps is a lot more than 190, but by the same logic the 8350K is also pounding itself. Drop the RAM to 3200 (probably CL16?) run -autoconfig, crank everything up to max and see how many fps are left. There's still the overclock and different scaling with RAM but at least you'll have a better idea of what the performance difference looks like.

That depends on the chip. And what you define as "excessive vcore".
I've seen 7600Ks at 5.0 with 1.25V.

Higher RAM speed is an Intel advantage in this case as even Zen 2's memory controller won't achieve as high of a RAM clock.

And with voltages I'm going off SiliconLottery's stats for how common certain clocks are.

Not really. There's 4400+ RAM on the QVLs. It's 2:1 for the IF clock beyond 3733 though so it only benefits bandwidth, not latency.
You asked for benchmarks with 3733, so now you don't get to act like 3200 vs 4133 is a completely fair and realistic scenario.

Either way different demos and different configs so I'm not sure where you're going with this. Or did you really think that just pushing it to 5.0 and using 4133CL17 would make the 8350K >50% faster than a 3600X?
The only other results (equally unfair) are actually higher than yours. So I don't get where the "pounding all these Zen2 chips" talk is coming from.

Clocks != voltage. 4.9 is less common than 4.8, 5.0 is less common than 4.9 and so on. Big surprise. That doesn't tell you what they consider excessive vcore or if they are limited by voltages at all, not thermals.
On top of that 8350Ks that make it to 5.2 by silicon lottery's standards exist and I can absolutely guarantee that those don't need excessive vcore for 4.9. I mean they claimed 82% to 4.9, 59% to 5.0. Does that mean they always use excessive vcore? That'd be your definition then, not theirs because they obviously think that vcore was fine.

Didn't realise people were using different benchmarks now, I just assumed everyone was posting benchmark1.dem results so there would be a common reference point.

And above 1.4v is excessive vcore for Coffee Lake because it's well beyond the optimal range of efficiency. Beyond 1.4v you start getting very large increase in heat generation for little improvement in clocks. Not unsafe (1.45+), but excessive.

So the first is benchmark1.dem on highest settings.
Moist's are completely different and the fps are higher than yours.

It's beyond me how you could've looked at this, read maybe 10% of the posts, and then concluded that your 8350K is way faster.

Vcore is just your opinion. Any oc lowers efficiency. The highest efficiency is at 0.8 to 0.9V iirc.

Waiting on Zen 2 benchmark1 run with mastercomfig v6 medium-low (or v7 low) then. Because if it can do at least 280fps under that condition, I will surely pick up a 3700X soon.

benchmark1 has outdated data, so it won't be fully accurate imo. We should move to a new one.

What's the most common one people are using now?

Well I have mine here: https://mega.nz/#!f8tlhDhR!nYgghqybOK15ObUykEczewB3242XHb_bJ4JP0rv1q6k
I guess some people have been using it already.

4812 frames 15.901 seconds 302.62 fps ( 3.30 ms/f) 40.062 fps variability

benchmark_test.dem

https://www.teamfortress.tv/post/910154/tf2-benchmarks

I get like 250 fps on that demo with R5 2600, the 8350k is cool and all but nowhere close to pounding Zen 2. Also you probably spent as much on Samsung B-die ram as your CPU, so its not that impressive.

My 3700X arrived today, I can certainly run such a benchmark for you (on 16GB dual channel 3000 MHz CL15) and I'll post the result tomorrow :)

should be trivial to oc the mem to 3200/3400

Probably so, but given the unstable nature of B450 boards I'll leave the memory alone until a somewhat-fully stable BIOS comes out. It's good enough that the XMP profile works.

for reference im getting 22256 frames 83.610 seconds 266.19 fps ( 3.76 ms/f) 21.324 fps variability
with dxlevel 81 mastercoms experimental on a 2700x with maxed out PBO undervolted on a NHD15 with 3333mhz cl16 mem with that demo

500 reboots later (not really, just under a hundred actually; don't do B450 boards, kids), here I am testing the 3700X. Results are in.

Setup details:
- Windows 10 1903
- GeForce driver 431.60
- GeForce Experience with replay enabled, no game optimization
- game details: mastercomfig v6 latest, 1080p, DX9, mastercomfig's launch parameters, mods: budhud, Less Distracting Explosions, mastercomfig addons (mouse tweaks, no pyroland, no soundscapes, no tutorial)
- tested with timedemo_runcount set at 3, so stuff being loaded in doesn't lower the results

Since there's a ton of data, I figured it'd be better to show it all in a table, and since TFTV doesn't have tables... Here's a Google Docs link: https://docs.google.com/spreadsheets/d/1AGtyRBvgcpYv2T8h9bSgx9IDd8fXFES175tPrsK5hMI/edit?usp=sharing

whitepuzzle, I think you'll be happy with the results ;)

ryzen 3600x with b350 board default cpu and 3200 memory
mastercomms low with no hats mod and no soundscape, pyroland, extra models etc.. dxlevel 81
with mastercomms demo benchmark_test
4812 frames 14.797 seconds 325.21 fps ( 3.07 ms/f) 70.159 fps variability

with the old benchmark1 demo from this thread
2639 frames 9.398 seconds 280.81 fps ( 3.56 ms/f) 26.269 fps variability

Is this dx8 or dx9?

DX9, launch parameters exact same from the mastercomfig's github page.

I was worried that Zen's higher memory latency compared to CFL would hurt TF2 performance (seeing how it's so sensitive to memory frequency and timings) but it looks like it doesn't matter too much.

At this rate, Zen 3 should be straight up faster than CFL, if not later batch Zen 2s with a more mature process and better clocks.

Well thanks to AMD not locking the cheaper chipsets down to 2666 MHz like Intel and people being aware that Zen needs a bit more than that to get decent latency it's not nearly as bad as with the Zen1 launch.

Would be great to see a comparision with 2667, 3200, 3733 and possibly higher though.

Maybe we'll get an N7+ (7nm + EUV) refresh similar to Zen+ (14->12nm which is just 14+) next year, that could be interesting.

According to roadmaps there is no Zen 2 refresh planned, it's straight on to Zen 3.

Which roadmaps? A refresh probably wouldn't be listed as architecture.
There's a lot of options. Zen3 is probably simply not be very different (Zen+ was literally identical except for the L2 fix so it couldn't be called Zen2), otherwise AMD wouldn't be able to release it just a year later. It is listed as "optimization" just like Zen+ for a reason. And just because Zen3 exists doesn't mean the whole Desktop lineup will get Zen3. E.g. no 2200, no 2400, 2300X and 2500X OEM only, no 2800X and Ryzen 3000 APUs are still Zen+. It's entirely possible for example that they'll do a die with more cores to bump up the per socket core count, but can't use that for the 6/8 core SKUs because it would mean disabling half the cores. Or that year desktop gets the short end of the stick and APUs get the new architecture first. EPYC only got 2 SKUs with the obligatory clock bump last year so after the big Zen2 update this year they might sit out next year again.

Let's just wait and see.