slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:31:08
(permalink)
arcky
Dude, I understand the concept you’re describing. I’m saying it’s entirely inapplicable to the data I presented. Even beyond the fact that in all the game benchmarks I posted I was 99% GPU bound, which immediately invalidates your main point regarding CpU bOTTLeNeCKinG, the primary focus of my analysis was comparative metrics, meaning RELATIVE scores between the two bios across synthetics and games, not ABSOLUTE scores. Please, if you’re not going to contribute meaningful content to the topic at hand, don’t muddy the discussion with superfluous criticism.
The problem with gaming might be temp related, as u increase PL with XC3 bios the cards getting hotter because time spent gaming is significantly more than one bench run. So in theory the gains from higher PL are lost in high temps which are resulting in lower boost because nvidia algo. Other possible reason might be just worse efficiency in games with xc3
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
arestavo
CLASSIFIED ULTRA Member
- Total Posts : 6801
- Reward points : 0
- Joined: 2008/02/06 06:58:57
- Location: Through the Scary Door
- Status: offline
- Ribbons : 77
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:33:59
(permalink)
arcky
Dude, I understand the concept you’re describing. I’m saying it’s entirely inapplicable to the data I presented. Even beyond the fact that in all the game benchmarks I posted I was 99% GPU bound, which immediately invalidates your main point regarding CpU bOTTLeNeCKinG, the primary focus of my analysis was comparative metrics, meaning RELATIVE scores between the two bios across synthetics and games, not ABSOLUTE scores. Please, if you’re not going to contribute meaningful content to the topic at hand, don’t muddy the discussion with superfluous criticism.
https://www.techpowerup.com/review/evga-geforce-rtx-3090-ftw3-ultra/32.html It's shown definitively in in-depth reviews - you get more performance increase at 2160p than you do at 1440p and 1080P. But that's all off topic as hell, and already proven.
|
slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:35:19
(permalink)
arestavo
arcky
Dude, I understand the concept you’re describing. I’m saying it’s entirely inapplicable to the data I presented. Even beyond the fact that in all the game benchmarks I posted I was 99% GPU bound, which immediately invalidates your main point regarding CpU bOTTLeNeCKinG, the primary focus of my analysis was comparative metrics, meaning RELATIVE scores between the two bios across synthetics and games, not ABSOLUTE scores. Please, if you’re not going to contribute meaningful content to the topic at hand, don’t muddy the discussion with superfluous criticism.
https://www.techpowerup.com/review/evga-geforce-rtx-3090-ftw3-ultra/32.html
It's shown definitively in in-depth reviews - you get more performance increase at 2160p than you do at 1440p and 1080P.
But that's all off topic as hell, and already proven.
man just let it be, its comparison of apples to apples, its valid, would be 4k better ? yes, does it matter now ? no
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:35:38
(permalink)
Thanks for providing a reasoned and on-topic reply slovak. I think both of your theories are valid, and just wanted to bring the data to the table to let others weigh in on possible reasonings or to provide their own data. I was all set to be content with the XC3 route until I actually looked at real-world performance beyond synthetics.
I will take a closer look at temps, but at first glance I believe my temps are lower on average in a game benchmark as the scenes are a bit more varied with more 'downtime' to reduce boosting and therefore temps, vs the synthetic benches which tend to present a constantly intense scene requiring consistently high boost clocks and higher temps. I'll take another look though to confirm.
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:39:43
(permalink)
arestavo
arcky
Dude, I understand the concept you’re describing. I’m saying it’s entirely inapplicable to the data I presented. Even beyond the fact that in all the game benchmarks I posted I was 99% GPU bound, which immediately invalidates your main point regarding CpU bOTTLeNeCKinG, the primary focus of my analysis was comparative metrics, meaning RELATIVE scores between the two bios across synthetics and games, not ABSOLUTE scores. Please, if you’re not going to contribute meaningful content to the topic at hand, don’t muddy the discussion with superfluous criticism.
https://www.techpowerup.com/review/evga-geforce-rtx-3090-ftw3-ultra/32.html
It's shown definitively in in-depth reviews - you get more performance increase at 2160p than you do at 1440p and 1080P.
But that's all off topic as hell, and already proven.
My man...I'm going to give you one more reply of my time to give you the benefit of the doubt that you may have just misunderstood my testing methodology. I'm comparing a 3090 FTW3 Ultra on an XC3 Bios, vs a 3090 FTW3 Ultra on the Beta '500w' XOC Bios. In all of my testing, the GPU is the bottleneck (99% GPU bound, this isn't a theory, it's a statistical result). Why in the name of Christ are you linking me relative performance gains of the 3090 over previous gen cards at various resolutions?
|
arestavo
CLASSIFIED ULTRA Member
- Total Posts : 6801
- Reward points : 0
- Joined: 2008/02/06 06:58:57
- Location: Through the Scary Door
- Status: offline
- Ribbons : 77
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:44:31
(permalink)
arcky
arestavo
arcky
Dude, I understand the concept you’re describing. I’m saying it’s entirely inapplicable to the data I presented. Even beyond the fact that in all the game benchmarks I posted I was 99% GPU bound, which immediately invalidates your main point regarding CpU bOTTLeNeCKinG, the primary focus of my analysis was comparative metrics, meaning RELATIVE scores between the two bios across synthetics and games, not ABSOLUTE scores. Please, if you’re not going to contribute meaningful content to the topic at hand, don’t muddy the discussion with superfluous criticism.
https://www.techpowerup.com/review/evga-geforce-rtx-3090-ftw3-ultra/32.html
It's shown definitively in in-depth reviews - you get more performance increase at 2160p than you do at 1440p and 1080P.
But that's all off topic as hell, and already proven.
My man...I'm going to give you one more reply of my time to give you the benefit of the doubt that you may have just misunderstood my testing methodology. I'm comparing a 3090 FTW3 Ultra on an XC3 Bios, vs a 3090 FTW3 Ultra on the Beta '500w' XOC Bios. In all of my testing, the GPU is the bottleneck (99% GPU bound, this isn't a theory, it's a statistical result). Why in the name of Christ are you linking me relative performance gains of the 3090 over previous gen cards at various resolutions?
bEcAuSe tHeSe CaRdS aRe NoT aS fAsT at 1080P and 1440P vs 2160P for an increase BECAUSE OF THE ARCHITECTURE OF THE GPU. Dead horse beaten for the 25th time now. If that's irrelevant to the information that you're trying to get across, then hell if I know because that's what I am seeing you try and push.
post edited by arestavo - 2020/11/18 09:48:02
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:50:54
(permalink)
Where in my analysis did I compare “1080P and 1440P vs 2160P” ? I think you’re fundamentally misunderstanding the data. Im comparing various synthetics to various synthetics, 1:1, apples to apples as Slovak said, and the same for the gaming benchmarks. I’m not comparing across resolutions. Purely to point out that RELATIVE gains from the XC3 bios are only evident in synthetic tests, not gaming. Many of those high-res synthetics are more cpu bound than my 1080p gaming benchmarks due to the complexity of the scenes+rtx etc.
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:54:12
(permalink)
Imagine looking at objective, unbiased DATA and concluding I’m trying to “push” something. I just ran 24 benchmarks to provide data to the community and EVGA for them to draw their own conclusions and hypothesis. I’m genuinely disheartened by the lack of critical thinking and aversion to math and science on display here.
|
arestavo
CLASSIFIED ULTRA Member
- Total Posts : 6801
- Reward points : 0
- Joined: 2008/02/06 06:58:57
- Location: Through the Scary Door
- Status: offline
- Ribbons : 77
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:56:27
(permalink)
arcky
Where in my analysis did I compare “1080P and 1440P vs 2160P” ? I think you’re fundamentally misunderstanding the data. Im comparing various synthetics to various synthetics, 1:1, apples to apples as Slovak said, and the same for the gaming benchmarks. I’m not comparing across resolutions. Purely to point out that RELATIVE gains from the XC3 bios are only evident in synthetic tests, not gaming. Many of those high-res synthetics are more cpu bound than my 1080p gaming benchmarks due to the complexity of the scenes+rtx etc.
Quite, and I'm sure they're all 4K too. And on top of that, I'm sure clockspeeds were normalized too. (Not just offsets)
|
slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 09:58:34
(permalink)
My last response the topic, From what I read I didnt get feeling hes trying to push anything, just basic 1:1 comparison. Again, Is higher res ideal for tests ? YES! But again it doesnt matter for this situation,I doubt gains on 4K would be much higher. BTW if u wanna have longer discussion, join my dc so we dont spam this thread with not so on-topic small talks.
post edited by slovak_killer - 2020/11/18 10:00:39
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
aukyo
New Member
- Total Posts : 19
- Reward points : 0
- Joined: 2018/09/24 05:03:10
- Status: offline
- Ribbons : 0
Flagged as Spam (1)
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 10:00:54
(permalink)
arcky are you a little mad?
|
slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 10:04:21
(permalink)
aukyo
arcky are you a little mad?
What value brings this question for others ? please dont spam this thread with personal issues.
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 10:04:22
(permalink)
aukyo
arcky are you a little mad?
(s)ad
|
aukyo
New Member
- Total Posts : 19
- Reward points : 0
- Joined: 2018/09/24 05:03:10
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 10:46:28
(permalink)
The (s)ad thing is your are mad because you didnt get the answer you wanted?
second what do we have benefit of your playing a game at 1080p?
|
PadinnPlays
New Member
- Total Posts : 53
- Reward points : 0
- Joined: 2020/10/07 23:56:48
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 11:14:30
(permalink)
|
jacoffey85
New Member
- Total Posts : 54
- Reward points : 0
- Joined: 2019/01/20 15:44:06
- Location: USA
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 11:29:10
(permalink)
arcky
Where in my analysis did I compare “1080P and 1440P vs 2160P” ? I think you’re fundamentally misunderstanding the data. Im comparing various synthetics to various synthetics, 1:1, apples to apples as Slovak said, and the same for the gaming benchmarks. I’m not comparing across resolutions. Purely to point out that RELATIVE gains from the XC3 bios are only evident in synthetic tests, not gaming. Many of those high-res synthetics are more cpu bound than my 1080p gaming benchmarks due to the complexity of the scenes+rtx etc.
You compared the different resolutions by using synthetic tests that were primarily higher resolutions and then comparing the gains to the gains of lower resolution gaming benchmarks for your conclusion. Did you not happen to notice in your results that Timespy (1440p Benchmark) is nearly half of a gain than the 4K benchmarks, and that the 8K benchmark had the highest gain? Using different resolutions potentially adds errors in the results and makes your data analysis fall under Simpson's Paradox.
HisR9-5950X || EVGA 3090 FTW3 || 32GB Corsair Dominator 4000CL16 || MSI Meg Ace X570 || Lian Li O11D XL || Custom Loop (EKWB/Optimus)
Hersi9-9900K || EVGA 2080Ti || 32GB Corsair Vengeance 3000CL14 || Asus Maximus Hero XI Z390 || Phanteks Evolv X || EKWB Custom Loop
|
slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 11:44:09
(permalink)
PadinnPlays
FYI Jacob replied to me on Twitter. They are watching our feedback.
https://twitter.com/EVGA_...29137109408440320?s=19
Nice, good to know someones watching feedback, Im willing to participate in whatever alpha/beta bios releases and give feedback, NDA is no problem tbh, we just want a better card right :)
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
jessmcca
New Member
- Total Posts : 1
- Reward points : 0
- Joined: 2020/09/30 09:04:27
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 12:04:25
(permalink)
Alright, I finally tested the XOC bios to see if I could hit the magic 500w number. With the default OC bios for the ftw3 3090 ultra I was able to pull 459w pretty consistently. With the XOC (OC bios only, I left the normal bios in place for the non-oc switch) flashed, I was able to hit 498.9w max and could sustain around 473w. So there is an improvement with the XOC bios but still not getting the consistent 500w pull (or 495w pull)
post edited by jessmcca - 2020/11/18 12:07:46
Attached Image(s)
|
orkan
New Member
- Total Posts : 75
- Reward points : 0
- Joined: 2020/09/24 06:00:54
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 12:23:40
(permalink)
Demonstrating performance of a video card by playing a game at 1080p is fully and completely retarded.
That is a fact.
|
changboy
iCX Member
- Total Posts : 258
- Reward points : 0
- Joined: 2020/04/16 16:36:52
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 13:10:04
(permalink)
Its normal to pull around 473w even less coz the 500w bios its not just 1 number of 500w
that bios have a target of 420w and can pull up to 500w depend of the load. Current varry like a roller coaster with graphik to simulate.
And the temperature of the card can limit the 420w target and the availability of this 500w.
Each time i begin play a game my card boost at 2100mhz then drop to 2070mhz then 2040mhz then 2010mhz then 1990mhz then 1980mhz.
If i have a waterblock, i think it will stay at 2100mhz and maybe i can oc a bit more to the 2150mhz and keep it stable.
So even with a 900w bios we will be always limited by temp and more you go high in MHZ more your temp need to be cold to be stable.
|
slovak_killer
Superclocked Member
- Total Posts : 135
- Reward points : 0
- Joined: 2020/11/09 00:21:28
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 13:18:58
(permalink)
changboy
Its normal to pull around 473w even less coz the 500w bios its not just 1 number of 500w
that bios have a target of 420w and can pull up to 500w depend of the load. Current varry like a roller coaster with graphik to simulate.
And the temperature of the card can limit the 420w target and the availability of this 500w.
Each time i begin play a game my card boost at 2100mhz then drop to 2070mhz then 2040mhz then 2010mhz then 1990mhz then 1980mhz.
If i have a waterblock, i think it will stay at 2100mhz and maybe i can oc a bit more to the 2150mhz and keep it stable.
So even with a 900w bios we will be always limited by temp and more you go high in MHZ more your temp need to be cold to be stable.
Well thats logical, 500W on air isnt sustainable, the thing is a lot of us wanna run custom water/subzero and we have no option to with 500W bios, on ln2 u can pull double the Wattage.
EXTREME OVERCLOCKER FROM SLOVAKIA my FTW3 related dc: discord.gg/kRSfJbvs9k
|
Xaxziminracs
Superclocked Member
- Total Posts : 103
- Reward points : 0
- Joined: 2014/04/20 12:10:38
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 13:34:50
(permalink)
|
sparetimepc
CLASSIFIED Member
- Total Posts : 2413
- Reward points : 0
- Joined: 2009/11/13 09:24:09
- Location: Altona, IL
- Status: offline
- Ribbons : 27
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 13:35:58
(permalink)
Yes this thread has gotten all out of hand just like so many others do get way off topic
|
cerealkeller
Superclocked Member
- Total Posts : 201
- Reward points : 0
- Joined: 2008/10/01 00:44:13
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 15:12:00
(permalink)
arcky
Imagine looking at objective, unbiased DATA and concluding I’m trying to “push” something. I just ran 24 benchmarks to provide data to the community and EVGA for them to draw their own conclusions and hypothesis. I’m genuinely disheartened by the lack of critical thinking and aversion to math and science on display here.
If it’s any consolation I thought your data was interesting, I like to see any relevant data and of course there are Exceptions, but I’d say all pertinent data may be relevant to somebody For a wide variety of reasons.
Intel i9 9900K 5.4 GHz Koolance CPU400 Water Block EVGA Z390 Dark 32 GB G Skill Trident 4000 MHz G Skill RAM Cooler EVGA RTX 3090 FTW3 Ultra (Optimus Block coming) Samsung 960 Evo 1 TB Seagate Fire Cuda 1 TB EVGA 1600 P2 PSU Pioneer BD-RW Thermaltake Core W200 Super Tower 2 x Alphacool Nexxus Monsta 480x90mm Rads, 280x65mm Phobya Rad, 240x55mm Koolance Rad 2 x Koolance PMP450S 24v Pumps, 2 x 770ml Koolance Reservoirs, 12 x Corsair ML140s, 4 x Corsair SP120s, 8 x Corsair HD120s, 8 x Cooler Master Sickle Flow 120s Windows 10
|
changboy
iCX Member
- Total Posts : 258
- Reward points : 0
- Joined: 2020/04/16 16:36:52
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 15:14:05
(permalink)
Do you play game sometime hehehe
|
kris1111
New Member
- Total Posts : 35
- Reward points : 0
- Joined: 2018/09/07 19:43:42
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 16:01:40
(permalink)
Not sure what scores are you guys getting in Port Royal, but this bios does completely nothing. I get max power draw at 440W regardless. The score on the standard BIOS is 14259 on the XOC 14273 Both on +160 and +1100, but the average clock is 1952MHz vs 1962MHz with an average temp of 33C. I see a lot of people hitting averages of 2120 to 2150MHz with much higher temps. Did I lose the silicon lottery?
post edited by kris1111 - 2020/11/18 16:08:18
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 16:02:29
(permalink)
People see 1080p and assume cpu bottleneck or high FPS. Without doing their own homework. It’s depressing. Go run the Watch Dogs Legion built in benchmark at 1080p with everything maxed. Everything. RTX full. No dlss. Guaranteed you’ll be sub 100fps average and GPU bound. If you don’t own WD do the same with SotTR, where you’ll have slightly higher FPS but still 99% GPU bound. I’ve made this extremely easy to reproduce, so I’ll await your input. Until then, your generalized responses are meaningless. And screenshot us your results pages for proof.
|
cerealkeller
Superclocked Member
- Total Posts : 201
- Reward points : 0
- Joined: 2008/10/01 00:44:13
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 16:08:26
(permalink)
this has been brought up before and I tested it for myself, but has anyone done any benchmarking with memory OCing only? The 3090s, and 3080s as well I would assume, have ECC memory which will add latency to your frametimes if it detects any errors. I tested the latency only, not the frame rate. But what I saw was huge latency spikes the higher the OC was, decreasing to certain extent as I lowered it, and the only clock speed where I saw no added latency was at 0, or stock. I'm just curious if anyone has some benchmarks to compare with just memory OC'd?
Intel i9 9900K 5.4 GHz Koolance CPU400 Water Block EVGA Z390 Dark 32 GB G Skill Trident 4000 MHz G Skill RAM Cooler EVGA RTX 3090 FTW3 Ultra (Optimus Block coming) Samsung 960 Evo 1 TB Seagate Fire Cuda 1 TB EVGA 1600 P2 PSU Pioneer BD-RW Thermaltake Core W200 Super Tower 2 x Alphacool Nexxus Monsta 480x90mm Rads, 280x65mm Phobya Rad, 240x55mm Koolance Rad 2 x Koolance PMP450S 24v Pumps, 2 x 770ml Koolance Reservoirs, 12 x Corsair ML140s, 4 x Corsair SP120s, 8 x Corsair HD120s, 8 x Cooler Master Sickle Flow 120s Windows 10
|
arcky
Superclocked Member
- Total Posts : 180
- Reward points : 0
- Joined: 2020/09/01 12:13:00
- Status: offline
- Ribbons : 0
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 16:10:23
(permalink)
cerealkeller
this has been brought up before and I tested it for myself, but has anyone done any benchmarking with memory OCing only? The 3090s, and 3080s as well I would assume, have ECC memory which will add latency to your frametimes if it detects any errors. I tested the latency only, not the frame rate. But what I saw was huge latency spikes the higher the OC was, decreasing to certain extent as I lowered it, and the only clock speed where I saw no added latency was at 0, or stock. I'm just curious if anyone has some benchmarks to compare with just memory OC'd?
Here’s my data with just memory OCs in PR, core at 0 offset: Memory 0: 13445 200: 13472 400: 13489 600: 13544 800: 13557 1000: 13569 1100: 13587 1150: 13590 1175: 13589 1200: 13588 1250: 13580 1300: 13546 1400: 13581 1600: crash
|
Nereus
Captain Goodvibes
- Total Posts : 18147
- Reward points : 0
- Joined: 2009/04/09 20:05:53
- Location: Brooklyn, NYC.
- Status: offline
- Ribbons : 58
Re: EVGA GeForce RTX 3090 FTW3 XOC BIOS BETA
2020/11/18 16:12:37
(permalink)
arcky
cerealkeller
this has been brought up before and I tested it for myself, but has anyone done any benchmarking with memory OCing only? The 3090s, and 3080s as well I would assume, have ECC memory which will add latency to your frametimes if it detects any errors. I tested the latency only, not the frame rate. But what I saw was huge latency spikes the higher the OC was, decreasing to certain extent as I lowered it, and the only clock speed where I saw no added latency was at 0, or stock. I'm just curious if anyone has some benchmarks to compare with just memory OC'd?
Here’s my data with just memory OCs in PR, core at 0 offset:
Memory
0: 13445
200: 13472
400: 13489
600: 13544
800: 13557
1000: 13569
1100: 13587
1150: 13590
1175: 13589
1200: 13588
1250: 13580
1300: 13546
1400: 13581
1600: crash
Thanks, that's very interesting seeing it start to drop after a certain point. Quite the juggling act to tune these cards in.
|