• 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.

Fallout 4 CPU benchmark thread(need some Zen3 and Zen4 results!)

bobharris

bobharris

bobharris

Associate
Joined
23 Jan 2023
Posts
4
Location
USA
bobharris said:
Man. It doesn't look like I bought myself much.

11700 -- 60
8000 -- 73

RX 6800 (Came from 1070....0 (zero) fps diff)
Ryzen 5 7600 (Came from 4770k.....20fps diff)?

XPG LANCER RGB DDR5 Desktop Memory: 32GB (2x16GB) 6000 MHz CL40-40-40​


My previous results.
11700 -- 40fps.
8000 -- 47fps.
DDR3/RX 6800/4770k.

The result seems kinda low.....
Click to expand...
My ram was NOT at 6000Mhz. It was at something considerably lower, like 4200Mhz, with who knows what kinda of timings.
More or less I've had an under clock on my ram since my new build for awhile now. Doing the F4 benches, seems to have brought it to light.
I did Cinebench single core to check. Closer to the 7600 than any other proc, so I'm good there. Now I'm running the XMP profile on that above ram correctly.

11700 -- 70fps
8000 -- 83fps
7600 Stock. Rx 6800 Stock. DDR5 6000Mhz CL40-40-40.

Not too shabby. Bought myself +10fps by just getting my ram speed/timings right. Previous result was under-clocked DDR5.
 
Last edited:

KompuKare

KompuKare

KompuKare

Associate
Joined
31 Dec 2010
Posts
2,441
Location
Sussex
bobharris said:
My ram was NOT at 6000Mhz. It was at something considerably lower, like 4200Mhz, with who knows what kinda of timings.
More or less I've had an under clock on my ram since my new build for awhile now. Doing the F4 benches, seems to have brought it to light.
I did Cinebench single core to check. Closer to the 7600 than any other proc, so I'm good there. Now I'm running the XMP profile on that above ram correctly.

11700 -- 70fps
8000 -- 83fps
7600 Stock. Rx 6800 Stock. DDR5 6000Mhz CL40-40-40.

Not too shabby. Bought myself +10fps by just getting my ram speed/timings right. Previous result was under-clocked DDR5.
Click to expand...
Well, in terms of rankings it moves you to #51 vs #63 of the score with the JEDEC DDR5. Your older Intel 4770K ranking was #76 (right at the bottom just above my 3570K with a RX 470).
hoIMAAD.png

As to why yours is still low? I'm afraid everything points to AMD's driver overhead for DX9/10/11 versus Nvidia (for DX12 things are largely reversed).
If we look at people who had run the benches with both GeForce and Radeon, we get a table like this:
YARp4E2.png

Which is all rather unfortunate as aside from Nvidia's last two '90 cards, I'd rather have the AMD cards due needing the VRAM (some Wabbajack Skyrim SE modlists actually recommend 16GB of VRAM and have a min of 12GB or so).

@bobharris without swapping your card for a GeForce we will never know, but your both average of 76.5 / 0.87 would get you around 88 which is top ten here.

Problem, I suspect, is that for the price of your 6800 you would only have got a 3070 whose 8GB is borderline so IMO the 6800 was the correct choice.
 

KompuKare

KompuKare

KompuKare

Associate
Joined
31 Dec 2010
Posts
2,441
Location
Sussex
bobharris said:
Quick thought. I do have a 1070 in the closet, my prev GPU. Would that give you another data point?
Click to expand...
It might be interesting if it's not too much trouble as we don't have any R5 7600 with Nvidia results.
I'm sure at actual playing resolutions there are probably big differences with the driver code of the 10x0 series and recent cards, but at low resolution like here it probably doesn't matter.
Thing is, it would only be academic as even with the overhead difference, I don't suppose you want to go back to an 8GB Pascal card!
 

KompuKare

KompuKare

KompuKare

Associate
Joined
31 Dec 2010
Posts
2,441
Location
Sussex
RavenXXX2 said:
You have the same score as my 13600k results mate.
Click to expand...
Opps, sorry. Copy and paste error, almost the same indeed!
Corrected it above, and here's the top 10:
71y9ten.png

So that's better. Number 4, probably without BLK overclocking like yuke and even faster memory it is hard to get much higher.

I keep including the S1/S2 ratio in the hope that it tells us something about drawcall vs NPCs (as CoreVega should have far more NPCs), but while there is variation I am unsure it means much.
 

CAT-THE-FIFTH

CAT-THE-FIFTH

CAT-THE-FIFTH

Soldato
OP
Joined
9 Nov 2009
Posts
24,845
Location
Planet Earth
KompuKare said:
Opps, sorry. Copy and paste error, almost the same indeed!
Corrected it above, and here's the top 10:
71y9ten.png

So that's better. Number 4, probably without BLK overclocking like yuke and even faster memory it is hard to get much higher.

I keep including the S1/S2 ratio in the hope that it tells us something about drawcall vs NPCs (as CoreVega should have far more NPCs), but while there is variation I am unsure it means much.
Click to expand...

It's outside and most of the NPCs are spread out in a large area - Diamond City has a lot of NPCs in a relative small area.
 

bookashade

bookashade

bookashade

Associate
Joined
25 Jun 2023
Posts
2
Location
Timbuktu
KompuKare said:
Probably close to one of the existing 13700KF or 13600K results:
Click to expand...
Is there any hope that at [email protected] GHz DDR4 3200CL14 I will get 100 FPS on Corvega? This may seem ridiculous to someone, but I need an upgrade specifically for Fallout 4. Otherwise, my 9900K suits me. And if so, is it better to look towards 7800x3D and 6000CL14 anyway?

I did a similar test 5 years ago on 3930K and 8700K.

Page 3 - Draw Call Performance In Fallout 4

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
forums.anandtech.com forums.anandtech.com

Page 3 - Draw Call Performance In Fallout 4

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
forums.anandtech.com forums.anandtech.com
I'll copy straight from there:

3930K @ 4.5 GHz / DDR3 @ 2400 MHz (9-11-11-31-1T) / GTX 980 Ti @ 1455/7900 MHz
Corvega: Draw Calls: 11704 / FPS: 60.9 / Diamond City: Draw Calls: 8025 / FPS: 72.3

8700K @ 4.8 GHz / 3000-14-15-15-30-T1 / GTX 980 Ti @ 1455/7900 MHz
Corvega: 81.6 FPS / Diamond City: 95.8 FPS

8700K @ 4.8 GHz / 3500-17-17-17-35-T2 / GTX 980 Ti @ 1455/7900 MHz
Corvega: 80.4 FPS / Diamond City: 95.7 FPS

8700K @ 4.8 GHz / DDR4 3000-17-17-17-35-T2 / GTX 980 Ti @ 1455/7900 MHz
Corvega: 75.9 FPS / Diamond City: 90.3 FPS

8700K @ 4 GHz / 3500-17-17-17-35-T2 / GTX 980 Ti @ 1455/7900 MHz
Corvega: 70.9 FPS / Diamond City: 83.5 FPS

But if I had 81 FPS on 8700K@3000CL14, wouldn't 100FPS be too low on 13th-gen?
 
Last edited:

Rob_B

Rob_B

Rob_B

Associate
Joined
29 Jan 2022
Posts
411
Location
UK
Damn, I was hoping to get some sort of trophy :D

I've actually tightened the RAM timings so it might even be faster than that now.
I wonder how using a discrete GPU could/would raise the score due to not having to hit the gfx part of the chip (see the 4700 vs 4650 results)

EDIT: Quick check with updated timings, could tighten more but I've not got the time or desire to mess any more :D

3400G w/Vega 11 - Stock
16GB (minus whatever for the APU) @3000mhz 18/17/17/45
Corvega - 36fps (11,685 dc)
Diamond City - 44.5fps (7,994 dc)
ENB v0.489
 
Last edited:

KompuKare

KompuKare

KompuKare

Associate
Joined
31 Dec 2010
Posts
2,441
Location
Sussex
mdaka42 said:
I don't get it. Why not just share the table via google spreadsheet? Using image for navigation is very inconvenient for obvious reasons.
Click to expand...
Well, Google sheets (and OpenOffice / Libre office Calc) aren't really able to do conditional formatting as well.
I did look at no registration required free online spreadsheet programs but they were even worse.
Google Sheets is far fancier but you have to sign up and can't use it anonymously AFAIK.
 
You must log in or register to reply here.
Back
Top Bottom