• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

HUB: Intel 13700K vs AMD 7800X3D

Using a 7800x3d Intel cant compete with the gaming experience cache does.
a game on intel cpu has to store info on mboard ram, 80ns.
AMD 7800x3d takes 7-8ns.
Intel isnt even in the game for gamers as they cant beat access time like that.
Intel simply can fix a 70ns or so delay.

I never had this type of fluid gaming experience as I do now with the 7800x3d.
placebo. I feel the same way with my laptop, never had such a great experience. Then I realize metrics don't lie, my laptop is much slower than my desktop, it's just my brain making things up. That's why we use independent tools called... Benchmarks.

It is a fact intel gives you a smoother experience. Lows are much much closer to averages on intel cpus, there is a much greater variation on x3ds.

The 3d works great as long as the game stays within the cache. When it doesn't we have what's now known as the amdip.
 
Last edited:
Decided to skip sleeping to test. So, the new map is playing "fine". 180 fps average (mostly gpu bound, card was 98% usage almost always), 1% lows at around 130. Last time we tested 1% lows were at 150-160 but whatever, not a big difference.

So then I decided to go back to Al mazrah. Holy cow, I was stuck at 140 fps averages with 1% lows dropping to 90ies. I remember I was getting close to 270+ FPS on this map, what happened?
Warzone 2 Season 4 Vondel - 345 Avg FPS 1% lows 196

Warzone 2 Season 4 Al Mazrah - 326 Avg FPS 1% lows 135
 
You are getting 350 on a 4090? How? Mine is pegged at 200 fps without dlss.
I can show you the way, but you may not like the answer... :cool:
Step 1 - Never, I repeat never, listen to Frame Chasers on anything AMD. :p
Step 2 - Buy a 7950X3D, and tune memory, FCLK and curve optimiser. Cheap Hynix M die is all you need, 5200Mhz speed or higher will do. No need to overclock the CPU using BCLK, stock will do. ;)
Step 3 - Result!
l3y4R4h.gif
 
Last edited:
I can show you the way, but you may not like the answer... :cool:
Step 1 - Never, I repeat never, listen to Frame Chasers on anything AMD. :p
Step 2 - Buy a 7950X3D, and tune memory, FCLK and curve optimiser. Cheap Hynix M die is all you need, 5200Mhz speed or higher will do. No need to overclock the CPU using BCLK, stock will do. ;)
Step 3 - Result!
l3y4R4h.gif
The card is at 98% usage man...it has nothing to do with the cpu, else dlss wouldn't help would it?

Frame chasers is irrelevant here last time we tested a stock 12900k had better lows than your 7950x 3d.
 
Last edited:
The card is at 98% usage man...it has nothing to do with the cpu, else dlss wouldn't help would it?

Frame chasers is irrelevant here last time we tested a stock 12900k had better lows than your 7950x 3d.
Strong doubt on the better lows part. Regardless you used DLSS so it was null and void as I ran native.

Game has got more demanding with the recent seasonal updates.

I notice that the 0.1% lows take a hammering whenever there is a scene transition now too. For example, when you die and it switches to a another player, or you enter gulag, etc.
 
Strong doubt on the better lows part. Regardless you used DLSS so it was null and void as I ran native.

Game has got more demanding with the recent seasonal updates.

I notice that the 0.1% lows take a hammering whenever there is a scene transition now too. For example, when you die and it switches to a another player, or you enter gulag, etc.
You can doubt but the videos are there, I still have them. You had around 80 to 100 higher averages but lower 1% lows. Dlss is irrelevant , if you are cpu bound then dlss doesn't increase your fps, lol.

Anwyas if a 4090 can get 350 fps in 1080p native then I guess the resolution scaling from I game doesn't work. I used 4k +50% scale instead of 1080p and the card was completely pegged, I'm gonna retry native now
 
Last edited:
The power figures are in line with what you should expect. In heavily multithreaded games, like open world + RT ala Cyberpunk, you'll see very high power usage because the CPU is juggling all the asset streaming (decompression); that's really the only way you'll ever see a high core count CPU actually put all of 'em to use in games, otherwise most of the work will fall onto 4 cores (+ medium load on 2 more). Similar thing happening with SotTR which is a game that also puts the CPU to use quite expertly. In something that's light on the CPU, and more GPU-focused, like RE, you can see a small difference because the CPU is mostly sleeping.

Shader compilation is also one of the few times a CPU will be put to work like it's almost in cinebench, so the big power draw in TloU is also normal, but we've seen how poorly the game does even in normal gameplay by keeping them fully loaded all the time for no real reason (other than dumb programming). Otherwise if it (the shader comp.) happens during the game you'll then see saturated cores and similar power-draw (see Spider-Man f.ex.).

In the future if you play games with DirectStorage w/ GPU decompression then the Intel CPUs should fall back in line vis-a-vis power draw during gameplay, because again only a few cores are really going to do heavy work in all games if asset decompression is off the table.

Overall AMD still completely clobbering Intel every which way, so hard to recommend anything blue. Nor do they look like they'll mount a comeback any time soon...
 
You can doubt but the videos are there, I still have them. You had around 80 to 100 higher averages but lower 1% lows. Dlss is irrelevant , if you are cpu bound then dlss doesn't increase your fps, lol.
DLSS is not irrelevant here, you just indicated your were GPU bound so of course its not irrelevant.
The card is at 98% usage man...it has nothing to do with the cpu, else dlss wouldn't help would it?
Once you've uploaded the two videos (1080P native - lowest settings) I can rerun it with DLSS enabled for a like for like comparison.

If you want you can place my config file in your COD Documents folder to make sure you use the same settings as me.
https://www.sendspace.com/file/gvfd0e

Anwyas if a 4090 can get 350 fps in 1080p native then I guess the resolution scaling from I game doesn't work. I used 4k +50% scale instead of 1080p and the card was completely pegged, I'm gonna retry native now
Stop tweaking, just use my config file and my settings.

This is what you did last time, got me to bench 1080P native then you used DLSS Ultra performance you cheeky sod. :cry:
 
Last edited:
DLSS is not irrelevant here, you just indicated your were GPU bound so of course its not irrelevant.
Exactly? I can't tell if you are serious or not. If the point is to test the CPUs, what difference does it make if I even drop to 240p? Anyways, im testing now at native 1080 instead of 4k with 50% scale, wish me luck
 
If you want you can place my config file in your COD Documents folder to make sure you use the same settings as me.
https://www.sendspace.com/file/gvfd0e


Stop tweaking, just use my config file and my settings.

This is what you did last time, got me to bench 1080P native then you used DLSS Ultra performance you cheeky sod. :cry:
It doesn't work, the game has the resolution locked at 4k, lol. The most have changed something
 
Last edited:
It doesn't work, the game has the resolution locked at 4k, lol. The most have changed something
I checked my config and I have the resolution, refresh rate and HDR set to Auto.

// Fullscreen Resolution
Resolution:0.0 = "Auto"

So it will use whatever is the default resolution of your display. Set to 1080P and launch the game.
 
Last edited:
Don't know either, but figured out what I was doing wrong. I had the game on ultra, that's why it was stuck at 98%. Never noticed, not im back to basic. Sorry :D
You have to be trolling me at this point. I want to see the settings in those videos. :cry:

Use my config file when you run the tests so our settings are identical!! @Bencher
 
Last edited:
Nope, tested TLOU with a power meter. Peaked at 491w, averaged at 475-480w. Hwunboxed shows 621w. Not very likely.

Possibly why you’re system is dropping 150fps to Matt with his 7950X3D :p

With Intel performance costs power, no way around that I’m afraid.
 
Possibly why you’re system is dropping 150fps to Matt with his 7950X3D :p

With Intel performance costs power, no way around that I’m afraid.
To be fair looks like he was using completely different settings, same as last time. :p

There shouldn't be such a large gap, so that's why its best to post a video like the ones I did for a proper comparison at the same settings.
 
You have to be trolling me at this point. I want to see the settings in those videos. :cry:

Use my config file when you run the tests so our settings are identical!! @Bencher
Οk copied your config file, seems to be working now. You have CAS enabled? Cause it has it enabled with your file.

FPS is a mess though, the new map is playing "fine", 198 fps average 130 1% lows, but Al mazrah is horrible, can't get over 140 fps average. Ill post videos later, but all in all im down 140 fps on averages compared to 6 months ago. They messed up something horribly
 
Last edited:
Back
Top Bottom