• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

12Gb GPUs already obsolete, brand new game takes up to 18Gb Vram at 1440p

Joined
Oct 15, 2011
Messages
2,581 (0.53/day)
Location
Springfield, Vermont
System Name KHR-1
Processor Ryzen 9 5900X
Motherboard ASRock B550 PG Velocita (UEFI-BIOS P3.40)
Memory 32 GB G.Skill RipJawsV F4-3200C16D-32GVR
Video Card(s) Sparkle Titan Arc A770 16 GB
Storage Western Digital Black SN850 1 TB NVMe SSD
Display(s) Alienware AW3423DWF OLED-ASRock PG27Q15R2A (backup)
Case Corsair 275R
Audio Device(s) Technics SA-EX140 receiver with Polk VT60 speakers
Power Supply eVGA Supernova G3 750W
Mouse Logitech G Pro (Hero)
Software Windows 11 Pro x64 23H2
WELP, RTX 4070 12gb just launched and a new game just rendered it useless as Star Wars Jedi Survivor reported to consume up to 18 gb vram at 1440p, jedi survivor is one of many upcoming games developed exclusively for current gen consoles which normally means very high vram requirements for their pc ports, thanks to consoles this time being equipped with generous 16gb ram, vram issues are continuing to get worse in light Nvidia shipping their midrange cards below 16gb vram, and it seemed like yesterday 12gb was considered to be the bare minimum, well maybe not anymore...


https://www.reddit.com/r/nvidia/comments/12zw867
- UPDATE : It appears that further testing shows the game eats up to 21Gb of Vram, PC ports of current gen games continues to render most GPUs today insufficient in light of gpu vendors giving you below minimum vram for anything but their top margin cards, is pc gaming dying?

View attachment 293490
Yikes! That makes nearly 98 percent of PCs look like a "potato"! Like the meme of emulating radically different hardware architectures on PCs!
With porting, it is more likely that the CPU gets hammered as well.
 
Last edited:
Joined
Oct 2, 2020
Messages
1,070 (0.67/day)
System Name ASUS TUF F15
Processor Intel Core i7-11800H
Motherboard ASUS FX506HC
Cooling Laptop built-in cooling lol
Memory 24 GB @ 3200
Video Card(s) Intel UHD & Nvidia RTX 3050 Mobile
Storage Adata XPG SX8200 Pro 512 GB
Display(s) Laptop built-in 144 Hz FHD screen
Audio Device(s) LOGITECH Z333 2.1-channel
Power Supply ASUS 180W PSU
Mouse Logitech G604
Keyboard Laptop built-in keyboard
Software Windows 10 Enterprise 20H2
Im really frustrated with how Backwards Nvida is going with their New GPUs with limited VRAM. I was looking forward to something more compatible for my CPU with the 4060ti coming out in may, but finding out they are releasing them with just 8GB of Vram! Heck the 3060s were released with 12Gb Vram, yet the new 4060 is pushed back to 8? I've been a fan of Nvidia but AMD seems to be taking the lead with more affordable GPUs with 16GB VRAM on the same tier GPU price wise. Because most of the modern games coming out the 8GB Vram card will be obsolete. Thanks Nvida. I'll pick the 16GB till you get your shyt together with the VRAM!!!
oh yeah, the 12 GB VRAM is the "POWER" factor of 3060, my ex 3060Ti with "just" 8 GB was like ~30% better @ 1440p just by "touch of magic".:D:roll:

I would absolutely get a used 3090 over a 4070 / ti - it's a no brainer in my book. As for dlss3, im really not a fan of it, and after having tested it in the different games, im not using it at all.



Are you unable to comprehend charts, or what?

It says ultra rt + no dlss = 11.6 gb vram consumption... ultra rt + dlss quality = 10.6 gb vram consumption.

Dlss does NOT increase vram consumption... it lowers the internal render res, so why on earth would it... -_-



Look at the OSD - it's not the gpu having issues, it's the cpu...
just don't forget to test properly the "used" 3090. Maybe, "Hynix-gate". maybe, "overheat-gate" or some other "surprise" will wait for you, naive boy...:D
 
Joined
Feb 23, 2019
Messages
6,208 (2.85/day)
Location
Poland
Processor Ryzen 7 5800X3D
Motherboard Gigabyte X570 Aorus Elite
Cooling Thermalright Phantom Spirit 120 SE
Memory 2x16 GB Crucial Ballistix 3600 CL16 Rev E @ 3600 CL14
Video Card(s) RTX3080 Ti FE
Storage SX8200 Pro 1 TB, Plextor M6Pro 256 GB, WD Blue 2TB
Display(s) LG 34GN850P-B
Case SilverStone Primera PM01 RGB
Audio Device(s) SoundBlaster G6 | Fidelio X2 | Sennheiser 6XX
Power Supply SeaSonic Focus Plus Gold 750W
Mouse Endgame Gear XM1R
Keyboard Wooting Two HE

That intro
 
  • Like
Reactions: 64K
Joined
Aug 30, 2006
Messages
7,228 (1.07/day)
System Name ICE-QUAD // ICE-CRUNCH
Processor Q6600 // 2x Xeon 5472
Memory 2GB DDR // 8GB FB-DIMM
Video Card(s) HD3850-AGP // FireGL 3400
Display(s) 2 x Samsung 204Ts = 3200x1200
Audio Device(s) Audigy 2
Software Windows Server 2003 R2 as a Workstation now migrated to W10 with regrets.
Good design. Good housekeeping. If this game had been optimally coded for PC, rather than a poorly coded console port with inefficient translation/interpretation layers, it would probably run at double the FPS with just 5GB GDDR5
 
Joined
May 8, 2021
Messages
247 (0.18/day)
Location
Scotland
System Name Scan 3XS System
Processor Intel Core i7 11700K
Motherboard ASUS TUF GAMING Z590-PLUS WIFI
Cooling Corsair Hydro H150i ELITE CAPELLIX RGB
Memory 64GB Kingston FURY Beast DDR4 3200Mhz
Video Card(s) 10GB EVGA GeForce RTX 3080 XC3 ULTRA GAMING
Storage 240GB EN605 M.2 SSD, 2TB Seagate HDD, 240GB Kingston SSD, 1TB Samsung 970 SSD
Display(s) Samsung C32JG5x Curved, 1080p, 1920 x 1080
Case Corsair iCUE 4000X RGB
Audio Device(s) Edifier G1500 RGB & Bluetooth Gaming 2.0 Speakers
Power Supply Corsair RM750x 80PLUS GOLD
Mouse Razer Mouse
Keyboard ROCCAT Horde
Software Microsoft Windows 11 Pro 64-Bit 24H2 [26100.3037]
Benchmark Scores Buying next week: Lexar NQ790 1TB SSD
Until they sort this problem out buying/playing this game will be a total washout and won't happen for me.
 

Mussels

Freshwater Moderator
Joined
Oct 6, 2004
Messages
58,413 (7.86/day)
Location
Oystralia
System Name Rainbow Sparkles (Power efficient, <350W gaming load)
Processor Ryzen R7 5800x3D (Undervolted, 4.45GHz all core)
Motherboard Asus x570-F (BIOS Modded)
Cooling Alphacool Apex UV - Alphacool Eisblock XPX Aurora + EK Quantum ARGB 3090 w/ active backplate
Memory 2x32GB DDR4 3600 Corsair Vengeance RGB @3866 C18-22-22-22-42 TRFC704 (1.4V Hynix MJR - SoC 1.15V)
Video Card(s) Galax RTX 3090 SG 24GB: Underclocked to 1700Mhz 0.750v (375W down to 250W))
Storage 2TB WD SN850 NVME + 1TB Sasmsung 970 Pro NVME + 1TB Intel 6000P NVME USB 3.2
Display(s) Phillips 32 32M1N5800A (4k144), LG 32" (4K60) | Gigabyte G32QC (2k165) | Phillips 328m6fjrmb (2K144)
Case Fractal Design R6
Audio Device(s) Logitech G560 | Corsair Void pro RGB |Blue Yeti mic
Power Supply Fractal Ion+ 2 860W (Platinum) (This thing is God-tier. Silent and TINY)
Mouse Logitech G Pro wireless + Steelseries Prisma XL
Keyboard Razer Huntsman TE ( Sexy white keycaps)
VR HMD Oculus Rift S + Quest 2
Software Windows 11 pro x64 (Yes, it's genuinely a good OS) OpenRGB - ditch the branded bloatware!
Benchmark Scores Nyooom.
16gb of shared RAM.
That shared RAM is GDDR6, it's VRAM the OS can also use.

The OS shrinks down to almost using none of it when gaming, giving the game enough to use stutter free.

That's max settings, no one needs, or gets any benefit, from max settings. That could be a exercise done with every card in the past.
I would say there is no real purpose from benchmarking at max settings, but it does help sell more, make more e-waste, and a planed obsolescence agenda. I don't get it.
There is a purpose, in that its fair and repeatable.

A better phrase would be there is no purpose to ONLY benchmarking at maximum settings, and all decent websites also benchmark medium settings as well (or, with things like RT off)
 
Joined
Nov 11, 2016
Messages
3,549 (1.18/day)
System Name The de-ploughminator Mk-III
Processor 9800X3D
Motherboard Gigabyte X870E Aorus Master
Cooling DeepCool AK620
Memory 2x32GB G.SKill 6400MT Cas32
Video Card(s) Asus RTX4090 TUF
Storage 4TB Samsung 990 Pro
Display(s) 48" LG OLED C4
Case Corsair 5000D Air
Audio Device(s) KEF LSX II LT speakers + KEF KC62 Subwoofer
Power Supply Corsair HX850
Mouse Razor Death Adder v3
Keyboard Razor Huntsman V3 Pro TKL
Software win11
All 16GB GDDR6 amounts to nothing when the game is poorly made anyways
IMG-2681.png

IMG-26810.png


so in 2023 we are gonna measure GPU performance by VRAM now?
 
Joined
Jul 15, 2020
Messages
1,036 (0.62/day)
System Name Dirt Sheep | Silent Sheep
Processor i5-2400 | 13900K (-0.02mV offset)
Motherboard Asus P8H67-M LE | Gigabyte AERO Z690-G, bios F29e Intel baseline
Cooling Scythe Katana Type 1 | Noctua NH-U12A chromax.black
Memory G-skill 2*8GB DDR3 | Corsair Vengeance 4*32GB DDR5 5200Mhz C40 @4000MHz
Video Card(s) Gigabyte 970GTX Mini | NV 1080TI FE (cap at 50%, 800mV)
Storage 2*SN850 1TB, 230S 4TB, 840EVO 128GB, WD green 2TB HDD, IronWolf 6TB, 2*HC550 18TB in RAID1
Display(s) LG 21` FHD W2261VP | Lenovo 27` 4K Qreator 27
Case Thermaltake V3 Black|Define 7 Solid, stock 3*14 fans+ 2*12 front&buttom+ out 1*8 (on expansion slot)
Audio Device(s) Beyerdynamic DT 990 (or the screen speakers when I'm too lazy)
Power Supply Enermax Pro82+ 525W | Corsair RM650x (2021)
Mouse Logitech Master 3
Keyboard Roccat Isku FX
VR HMD Nop.
Software WIN 10 | WIN 11
Benchmark Scores CB23 SC: i5-2400=641 | i9-13900k=2325-2281 MC: i5-2400=i9 13900k SC | i9-13900k=37240-35500
People's behavior around Vram always been quite dramatic, just as a system behave (read- preform) at a given Vram.
It is a binary process with dramatic 'step function' (non linear) behavior, contrary to many other graphic parameters effect that are gradual. You either have enough or not. No middle ground in between.
But broken games and misunderstanding memory allocation drove some out of their loop..
 
Joined
May 17, 2021
Messages
3,421 (2.51/day)
Processor Ryzen 5 5700x
Motherboard B550 Elite
Cooling Thermalright Perless Assassin 120 SE
Memory 32GB Fury Beast DDR4 3200Mhz
Video Card(s) Gigabyte 3060 ti gaming oc pro
Storage Samsung 970 Evo 1TB, WD SN850x 1TB, plus some random HDDs
Display(s) LG 27gp850 1440p 165Hz 27''
Case Lian Li Lancool II performance
Power Supply MSI 750w
Mouse G502
There is a purpose, in that its fair and repeatable.

A better phrase would be there is no purpose to ONLY benchmarking at maximum settings, and all decent websites also benchmark medium settings as well (or, with things like RT off)

it's a game review. Reviewing at max settings, low settings, medium settings, ultra settings, are all fair and repeatable, just select the option on the menu. What are you talking about?
 
Joined
Sep 17, 2014
Messages
23,179 (6.10/day)
Location
The Washing Machine
System Name Tiny the White Yeti
Processor 7800X3D
Motherboard MSI MAG Mortar b650m wifi
Cooling CPU: Thermalright Peerless Assassin / Case: Phanteks T30-120 x3
Memory 32GB Corsair Vengeance 30CL6000
Video Card(s) ASRock RX7900XT Phantom Gaming
Storage Lexar NM790 4TB + Samsung 850 EVO 1TB + Samsung 980 1TB + Crucial BX100 250GB
Display(s) Gigabyte G34QWC (3440x1440)
Case Lian Li A3 mATX White
Audio Device(s) Harman Kardon AVR137 + 2.1
Power Supply EVGA Supernova G2 750W
Mouse Steelseries Aerox 5
Keyboard Lenovo Thinkpad Trackpoint II
VR HMD HD 420 - Green Edition ;)
Software W11 IoT Enterprise LTSC
Benchmark Scores Over 9000
it's a game review. Reviewing at max settings, low settings, medium settings, ultra settings, are all fair and repeatable, just select the option on the menu. What are you talking about?
Not true at all, something as simple as a patch for the game can make graphics presets act differently, whereas the 'highest setting' will always be the highest. This also means you can judge the worst case perf scenario, instead of best case, or 'somewhat decent' or whatever abstract thing you want to call it.

I think resolution on its own is a pretty good differentiator, and having three resolutions tested each on ultra gives you a great bandwidth of expected performance.
 
Last edited:
Joined
Jun 1, 2011
Messages
4,827 (0.97/day)
Location
in a van down by the river
Processor faster at instructions than yours
Motherboard more nurturing than yours
Cooling frostier than yours
Memory superior scheduling & haphazardly entry than yours
Video Card(s) better rasterization than yours
Storage more ample than yours
Display(s) increased pixels than yours
Case fancier than yours
Audio Device(s) further audible than yours
Power Supply additional amps x volts than yours
Mouse without as much gnawing as yours
Keyboard less clicky than yours
VR HMD not as odd looking as yours
Software extra mushier than yours
Benchmark Scores up yours
That shared RAM is GDDR6, it's VRAM the OS can also use.

The OS shrinks down to almost using none of it when gaming, giving the game enough to use stutter free.

We talking consoles? I believe the OS has its own pool of 512MB DDR4 ram. The 16GB is shared between video and the game engine system RAM if I recall but I don't pay too much attention to the consoles so maybe it changed.
 
Joined
May 17, 2021
Messages
3,421 (2.51/day)
Processor Ryzen 5 5700x
Motherboard B550 Elite
Cooling Thermalright Perless Assassin 120 SE
Memory 32GB Fury Beast DDR4 3200Mhz
Video Card(s) Gigabyte 3060 ti gaming oc pro
Storage Samsung 970 Evo 1TB, WD SN850x 1TB, plus some random HDDs
Display(s) LG 27gp850 1440p 165Hz 27''
Case Lian Li Lancool II performance
Power Supply MSI 750w
Mouse G502
Not true at all, something as simple as a patch for the game can make graphics presets act differently, whereas the 'highest setting' will always be the highest. This also means you can judge the worst case perf scenario, instead of best case, or 'somewhat decent' or whatever abstract thing you want to call it.

I think resolution on its own is a pretty good differentiator, and having three resolutions tested each on ultra gives you a great bandwidth of expected performance.

You test a game on a specific time, a day later max settings can act differently with a patch, i don't see any difference there. As long as all the cards were tested with the same game version that is irrelevant.
 

Mussels

Freshwater Moderator
Joined
Oct 6, 2004
Messages
58,413 (7.86/day)
Location
Oystralia
System Name Rainbow Sparkles (Power efficient, <350W gaming load)
Processor Ryzen R7 5800x3D (Undervolted, 4.45GHz all core)
Motherboard Asus x570-F (BIOS Modded)
Cooling Alphacool Apex UV - Alphacool Eisblock XPX Aurora + EK Quantum ARGB 3090 w/ active backplate
Memory 2x32GB DDR4 3600 Corsair Vengeance RGB @3866 C18-22-22-22-42 TRFC704 (1.4V Hynix MJR - SoC 1.15V)
Video Card(s) Galax RTX 3090 SG 24GB: Underclocked to 1700Mhz 0.750v (375W down to 250W))
Storage 2TB WD SN850 NVME + 1TB Sasmsung 970 Pro NVME + 1TB Intel 6000P NVME USB 3.2
Display(s) Phillips 32 32M1N5800A (4k144), LG 32" (4K60) | Gigabyte G32QC (2k165) | Phillips 328m6fjrmb (2K144)
Case Fractal Design R6
Audio Device(s) Logitech G560 | Corsair Void pro RGB |Blue Yeti mic
Power Supply Fractal Ion+ 2 860W (Platinum) (This thing is God-tier. Silent and TINY)
Mouse Logitech G Pro wireless + Steelseries Prisma XL
Keyboard Razer Huntsman TE ( Sexy white keycaps)
VR HMD Oculus Rift S + Quest 2
Software Windows 11 pro x64 (Yes, it's genuinely a good OS) OpenRGB - ditch the branded bloatware!
Benchmark Scores Nyooom.
it's a game review. Reviewing at max settings, low settings, medium settings, ultra settings, are all fair and repeatable, just select the option on the menu. What are you talking about?
Multiple things, we're on page 12 here - try and keep up.


Running out of VRAM has less of a performance impact on a high end PC - because the CPU and RAM can feed new content to the GPU fast enough to not stutter.

Someone with DDR4 2133 or a SATA drive would have an entirely nasty experience when they run out of VRAM, vs w1zzards DDR5 9001 and NVME
 
Joined
Sep 17, 2014
Messages
23,179 (6.10/day)
Location
The Washing Machine
System Name Tiny the White Yeti
Processor 7800X3D
Motherboard MSI MAG Mortar b650m wifi
Cooling CPU: Thermalright Peerless Assassin / Case: Phanteks T30-120 x3
Memory 32GB Corsair Vengeance 30CL6000
Video Card(s) ASRock RX7900XT Phantom Gaming
Storage Lexar NM790 4TB + Samsung 850 EVO 1TB + Samsung 980 1TB + Crucial BX100 250GB
Display(s) Gigabyte G34QWC (3440x1440)
Case Lian Li A3 mATX White
Audio Device(s) Harman Kardon AVR137 + 2.1
Power Supply EVGA Supernova G2 750W
Mouse Steelseries Aerox 5
Keyboard Lenovo Thinkpad Trackpoint II
VR HMD HD 420 - Green Edition ;)
Software W11 IoT Enterprise LTSC
Benchmark Scores Over 9000
You test a game on a specific time, a day later max settings can act differently with a patch, i don't see any difference there. As long as all the cards were tested with the same game version that is irrelevant.
What about all the checkboxes that arent directly linked to presets? What about games that havent got presets? Why then choose one over another graphics option? All these problems are gone if you simply use everything the game has.
 
Joined
May 17, 2021
Messages
3,421 (2.51/day)
Processor Ryzen 5 5700x
Motherboard B550 Elite
Cooling Thermalright Perless Assassin 120 SE
Memory 32GB Fury Beast DDR4 3200Mhz
Video Card(s) Gigabyte 3060 ti gaming oc pro
Storage Samsung 970 Evo 1TB, WD SN850x 1TB, plus some random HDDs
Display(s) LG 27gp850 1440p 165Hz 27''
Case Lian Li Lancool II performance
Power Supply MSI 750w
Mouse G502
Multiple things, we're on page 12 here - try and keep up.


Running out of VRAM has less of a performance impact on a high end PC - because the CPU and RAM can feed new content to the GPU fast enough to not stutter.

Someone with DDR4 2133 or a SATA drive would have an entirely nasty experience when they run out of VRAM, vs w1zzards DDR5 9001 and NVME

if you have a high end PC with DDR5 9001 you're probably not using a card with 8GB
 
Joined
Sep 17, 2014
Messages
23,179 (6.10/day)
Location
The Washing Machine
System Name Tiny the White Yeti
Processor 7800X3D
Motherboard MSI MAG Mortar b650m wifi
Cooling CPU: Thermalright Peerless Assassin / Case: Phanteks T30-120 x3
Memory 32GB Corsair Vengeance 30CL6000
Video Card(s) ASRock RX7900XT Phantom Gaming
Storage Lexar NM790 4TB + Samsung 850 EVO 1TB + Samsung 980 1TB + Crucial BX100 250GB
Display(s) Gigabyte G34QWC (3440x1440)
Case Lian Li A3 mATX White
Audio Device(s) Harman Kardon AVR137 + 2.1
Power Supply EVGA Supernova G2 750W
Mouse Steelseries Aerox 5
Keyboard Lenovo Thinkpad Trackpoint II
VR HMD HD 420 - Green Edition ;)
Software W11 IoT Enterprise LTSC
Benchmark Scores Over 9000
if you have a high end PC with DDR5 9001 you're probably not using a card with 8GB
Ehhh exactly, you use it on the PC with slower memory and CPU and so you will meet stutter more readily on 8GB.
 
Joined
May 17, 2021
Messages
3,421 (2.51/day)
Processor Ryzen 5 5700x
Motherboard B550 Elite
Cooling Thermalright Perless Assassin 120 SE
Memory 32GB Fury Beast DDR4 3200Mhz
Video Card(s) Gigabyte 3060 ti gaming oc pro
Storage Samsung 970 Evo 1TB, WD SN850x 1TB, plus some random HDDs
Display(s) LG 27gp850 1440p 165Hz 27''
Case Lian Li Lancool II performance
Power Supply MSI 750w
Mouse G502
What about all the checkboxes that arent directly linked to presets? What about games that havent got presets? Why then choose one over another graphics option? All these problems are gone if you simply use everything the game has.

what if they add more options with an update. or change the options. it happens a lot.
what do check all the boxes mean, vsync? motion blur? usually the presets are linked to the relevant options. And whatever you choose for max settings that you find relevant you can choose for any other preset, i'm sure it's not just checking all the boxes you find, some are irrelevant, some make things worst.

what if's... max settings brings you no more security then any other option.
 

Mussels

Freshwater Moderator
Joined
Oct 6, 2004
Messages
58,413 (7.86/day)
Location
Oystralia
System Name Rainbow Sparkles (Power efficient, <350W gaming load)
Processor Ryzen R7 5800x3D (Undervolted, 4.45GHz all core)
Motherboard Asus x570-F (BIOS Modded)
Cooling Alphacool Apex UV - Alphacool Eisblock XPX Aurora + EK Quantum ARGB 3090 w/ active backplate
Memory 2x32GB DDR4 3600 Corsair Vengeance RGB @3866 C18-22-22-22-42 TRFC704 (1.4V Hynix MJR - SoC 1.15V)
Video Card(s) Galax RTX 3090 SG 24GB: Underclocked to 1700Mhz 0.750v (375W down to 250W))
Storage 2TB WD SN850 NVME + 1TB Sasmsung 970 Pro NVME + 1TB Intel 6000P NVME USB 3.2
Display(s) Phillips 32 32M1N5800A (4k144), LG 32" (4K60) | Gigabyte G32QC (2k165) | Phillips 328m6fjrmb (2K144)
Case Fractal Design R6
Audio Device(s) Logitech G560 | Corsair Void pro RGB |Blue Yeti mic
Power Supply Fractal Ion+ 2 860W (Platinum) (This thing is God-tier. Silent and TINY)
Mouse Logitech G Pro wireless + Steelseries Prisma XL
Keyboard Razer Huntsman TE ( Sexy white keycaps)
VR HMD Oculus Rift S + Quest 2
Software Windows 11 pro x64 (Yes, it's genuinely a good OS) OpenRGB - ditch the branded bloatware!
Benchmark Scores Nyooom.
if you have a high end PC with DDR5 9001 you're probably not using a card with 8GB
Totally missed the point there.
W1zzards results show little change to running out of VRAM, while home users with systems that have slower components might have serious issues.


The problem is that testing takes a lot of time, and games have dozens of settings leading to hundreds or thousands of combinations that are impossible to test all of them, especially when you add in external factors like driver settings and nvidia/AMD control panel settings (Hence w1zzard simply not installing them)
 
Joined
Mar 22, 2020
Messages
27 (0.02/day)
Multiple things, we're on page 12 here - try and keep up.


Running out of VRAM has less of a performance impact on a high end PC - because the CPU and RAM can feed new content to the GPU fast enough to not stutter.

Someone with DDR4 2133 or a SATA drive would have an entirely nasty experience when they run out of VRAM, vs w1zzards DDR5 9001 and NVME
Not really no, the limiting factor will be the PCIe link speed and latency. Which is one or more order of magnitude higher than vram. Which means no matter the system if you run out of vram the experience will be comparably terrible.

Also the initial article proved to be bollocks for the actual release.

Also measuring vram occupation in ultra on a 24GB card with an engine that adapts its use to the available vram is pointless to assess vram needs on lesser modes and cards.

Also consoles have 16GB unified memory, but about 4GB are reserved for the system, so that's 12GB ram + vram. And when you run out of it you hard crash. So Games will be much more memory constrained on consoles than what is usually believed. Situation is even worse on the series s which has about 8GB available memory.
 
Joined
Sep 1, 2020
Messages
2,553 (1.58/day)
Location
Bulgaria
Not really no, the limiting factor will be the PCIe link speed and latency
I expect this bottleneck to improve a little with the next generation of graphics cards, which finally will be PCIe 5.0, unlike today's generation, which unfortunately missed this.
 
Joined
May 17, 2021
Messages
3,421 (2.51/day)
Processor Ryzen 5 5700x
Motherboard B550 Elite
Cooling Thermalright Perless Assassin 120 SE
Memory 32GB Fury Beast DDR4 3200Mhz
Video Card(s) Gigabyte 3060 ti gaming oc pro
Storage Samsung 970 Evo 1TB, WD SN850x 1TB, plus some random HDDs
Display(s) LG 27gp850 1440p 165Hz 27''
Case Lian Li Lancool II performance
Power Supply MSI 750w
Mouse G502
Totally missed the point there.
W1zzards results show little change to running out of VRAM, while home users with systems that have slower components might have serious issues.


The problem is that testing takes a lot of time, and games have dozens of settings leading to hundreds or thousands of combinations that are impossible to test all of them, especially when you add in external factors like driver settings and nvidia/AMD control panel settings (Hence w1zzard simply not installing them)

there are A LOT of systems with 4090's/4080's combined with a cheaper cpu and motherboard, and ram,etc... or 60 class cards with even lower class cpu/ram/... it happens a lot, in fact that's the rule everyone used for decades to build their own PC's. High end mobos, ram and a cheap gt1030, that you don't find a lot.
People were using 3700k's with 2080tis and shit like that until very recently.
Those assumptions that the PC is one homogenous block is very wrong.
 
Joined
Mar 22, 2020
Messages
27 (0.02/day)
I expect this bottleneck to improve a little with the next generation of graphics cards, which finally will be PCIe 5.0, unlike today's generation, which unfortunately missed this.
Bandwidth is x2 and latency is not expected to be that different from what I can read so this will not change the situation dramatically. That was my point actually, pcie performance is so low compared to ddr and worse gddr that no matter the quality of your setup for the same amount of missing vram you'll basically have the same experience.
 
Joined
Mar 10, 2010
Messages
11,880 (2.18/day)
Location
Manchester uk
System Name RyzenGtEvo/ Asus strix scar II
Processor Amd R5 5900X/ Intel 8750H
Motherboard Crosshair hero8 impact/Asus
Cooling 360EK extreme rad+ 360$EK slim all push, cpu ek suprim Gpu full cover all EK
Memory Gskill Trident Z 3900cas18 32Gb in four sticks./16Gb/16GB
Video Card(s) Asus tuf RX7900XT /Rtx 2060
Storage Silicon power 2TB nvme/8Tb external/1Tb samsung Evo nvme 2Tb sata ssd/1Tb nvme
Display(s) Samsung UAE28"850R 4k freesync.dell shiter
Case Lianli 011 dynamic/strix scar2
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup, corsair void pro headset
Power Supply corsair 1200Hxi/Asus stock
Mouse Roccat Kova/ Logitech G wireless
Keyboard Roccat Aimo 120
VR HMD Oculus rift
Software Win 10 Pro
Benchmark Scores laptop Timespy 6506
Not really no, the limiting factor will be the PCIe link speed and latency. Which is one or more order of magnitude higher than vram. Which means no matter the system if you run out of vram the experience will be comparably terrible.

Also the initial article proved to be bollocks for the actual release.

Also measuring vram occupation in ultra on a 24GB card with an engine that adapts its use to the available vram is pointless to assess vram needs on lesser modes and cards.

Also consoles have 16GB unified memory, but about 4GB are reserved for the system, so that's 12GB ram + vram. And when you run out of it you hard crash. So Games will be much more memory constrained on consoles than what is usually believed. Situation is even worse on the series s which has about 8GB available memory.
A console crashes when hitting a Vram issue?!?.
I smell something real bad.

Never seen that as a many year and present console user(ps5)..

The point your forgetting is a console IS VERY optimised.
A pc is not.

The 8GB YOU say a console has it uses in a very pre determined, optimised way to pass you med to high IQ ,, specifically chosen for a acceptable performance level.

Pc doesn't have that, it has at least 3 minimum ,present render paths to choose, many other API and a settings page which is far and away beyond what a console can do at all in some cases but in all cases that exceeds the console IQ.

A consoles hardware is known, little Johnny on his 3070Ti doesn't often Know what he has, very often miss configures his shit as evidenced daily here, and his hardware was randomly chosen usually on BOM cost.

And finally a console has tech a pc doesn't, it also doesn't have the RGB monitoring, driver update checking, total f£#@££ tat random software that a pc also runs concurrently .
 

AsRock

TPU addict
Joined
Jun 23, 2007
Messages
19,172 (2.98/day)
Location
UK\USA
Processor AMD 3900X \ AMD 7700X
Motherboard ASRock AM4 X570 Pro 4 \ ASUS X670Xe TUF
Cooling D15
Memory Patriot 2x16GB PVS432G320C6K \ G.Skill Flare X5 F5-6000J3238F 2x16GB
Video Card(s) eVga GTX1060 SSC \ XFX RX 6950XT RX-695XATBD9
Storage Sammy 860, MX500, Sabrent Rocket 4 Sammy Evo 980 \ 1xSabrent Rocket 4+, Sammy 2x990 Pro
Display(s) Samsung 1080P \ LG 43UN700
Case Fractal Design Pop Air 2x140mm fans from Torrent \ Fractal Design Torrent 2 SilverStone FHP141x2
Audio Device(s) Yamaha RX-V677 \ Yamaha CX-830+Yamaha MX-630 \Paradigm 7se MKII, Paradigm 5SE MK1 , Blue Yeti
Power Supply Seasonic Prime TX-750 \ Corsair RM1000X Shift
Mouse Steelseries Sensei wireless \ Steelseries Sensei wireless
Keyboard Logitech K120 \ Wooting Two HE
Benchmark Scores Meh benchmarks.
Its clear as day to me that we're moving on from 8GB. I saw it in person, I experienced it on an 8GB card. I've moved to a 20GB card now and again I see the major differences, and not just in FPS. 8GB is history as much as 3-4 GB were in 2021. 6GB right now is definitely exit as well... 1060~2060(S) is done.

Can you cope with 8GB, sure, but you're coping. If that's on an already owned card, all is well. If its on anything you buy today, you're a fucking idiot and in denial. That's the long and short of it. And beyond 8GB it still all depends on your future upgrade plans imho if, say, 12GB is sufficient. But the reality is, a 4070 or 4070ti with 12GB has way more core perf than you get in relative VRAM, and again... everybody is at liberty to think their thing of it, but it is what it is, the numbers just don't lie, and no matter how much magical (cache) sauce you sprinkle over games, lacking any kind of RAM capacity is a performance killer that will rear its head and force you to an upgrade because shit didn't just get slow, it got unplayable.

It is exactly like you say with CPU cores/threads. Have enough, you're good. Have more? Pointless. Have too little? You're screwed.

Well it's common sense that vram amount is going to go up over time, like OMG my 1st PC had 2MB up graded with 1/2MB ram chips.

How ever the starwars example used in this thread is a terrible one, like come on it don't even look all that great, just a pile of badly optimized crap.

And think it's around 12-16GB is were the sweet spot going to be, nVidia lost a sale with me due to it.

One thing for sure 6-8GB v cards should be avoided.
 
Joined
Mar 22, 2020
Messages
27 (0.02/day)
A console crashes when hitting a Vram issue?!?.
I smell something real bad.
Indeed it would be really bad if it happened on a consumer console, which is why developers avoid this at all costs. But the limit exists and is even more stringent than on PCs.

The 8GB YOU say a console has it uses in a very pre determined, optimised way to pass you med to high IQ ,, specifically chosen for a acceptable performance level.

Pc doesn't have that, it has at least 3 minimum ,present render paths to choose, many other API and a settings page which is far and away beyond what a console can do at all in some cases but in all cases that exceeds the console IQ.

A consoles hardware is known, little Johnny on his 3070Ti doesn't often Know what he has, very often miss configures his shit as evidenced daily here, and his hardware was randomly chosen usually on BOM cost.
And yes a console is a single hardware and software combination but this fact really doesn't do much for the vram usage : between a dx12 xbox and a PC, for identical graphics settings the needs also are basically the same. The main difference here is console are more efficient than mid range PCs at streaming data from the drive. The ps5 is more efficient than anything I know all things considered.

And finally a console has tech a pc doesn't, it also doesn't have the RGB monitoring, driver update checking, total f£#@££ tat random software that a pc also runs concurrently .
Yes that's true, people often forget other software can occupy your vram. That's particularly true for modern browsers that use a lot (up to several GB) of it and don't seem to back down when a game is launched that would need it. The infamous hardware unboxed video was an hilarious example of this: the first game tested (and one of the few where they bothered actually measuring the game's usage) showed 4-5GB of other software filling the 8GB of 3070 vram...
 
Top