-
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.
- Status
Associate
"Lol" was literally going to be my only response to what do I think about the 4070ti... But you put it better xD
Caporegime
The RTX4070TI isn't a 70 series card but a rebadged 60/60TI series card. The GPU is under 300MM2 too. People keep forgetting Nvidia released an AD103,so the AD104 actually is now the same position as the GA106 in the lineup and the AD103 is basically where the GA104 was positioned.
But even ignoring die size,just look at the shader count:
NVIDIA GeForce RTX 4090 Specs
NVIDIA AD102, 2520 MHz, 16384 Cores, 512 TMUs, 176 ROPs, 24576 MB GDDR6X, 1313 MHz, 384 bitwww.techpowerup.com
NVIDIA GeForce RTX 4070 Ti Specs
NVIDIA AD104, 2610 MHz, 7680 Cores, 240 TMUs, 80 ROPs, 12288 MB GDDR6X, 1313 MHz, 192 bitwww.techpowerup.com
The RTX4070TI has 47% of the shader count of the RTX4090,half the VRAM capacity and 50% of the memory bandwith.
Now look at the RTX3090TI vs the RTX3070TI and RTX3060:
NVIDIA GeForce RTX 3090 Ti Specs
NVIDIA GA102, 1860 MHz, 10752 Cores, 336 TMUs, 112 ROPs, 24576 MB GDDR6X, 1313 MHz, 384 bit
NVIDIA GeForce RTX 3070 Ti Specs
NVIDIA GA104, 1770 MHz, 6144 Cores, 192 TMUs, 96 ROPs, 8192 MB GDDR6X, 1188 MHz, 256 bitwww.techpowerup.com
NVIDIA GeForce RTX 3060 12 GB Specs
NVIDIA GA106, 1777 MHz, 3584 Cores, 112 TMUs, 48 ROPs, 12288 MB GDDR6, 1875 MHz, 192 bit
The RTX3070TI has 57% of the shader count of the RTX3090TI and 60.3% of the memory bandwidth of the RTX3090TI.
The RTX3060 has 33% of the shader count of the RTX3090TI and 35.7% of the memory bandwidth of the RTX3090TI.
Now look at how it is compared to the RTX3060TI:
NVIDIA GeForce RTX 3060 Ti Specs
NVIDIA GA104, 1665 MHz, 4864 Cores, 152 TMUs, 80 ROPs, 8192 MB GDDR6, 1750 MHz, 256 bit
It has 45.2% of the shader count of the RTX3090TI and 44% of the memory bandwidth.
So the RTX4070TI essentially has a dGPU chip about the same of a 106 series chip,but relative performance to the top is closer to a 60TI/60 Super class relative performance.
The RTX4070TI relative to the RTX4090,is basically what the RTX3060TI was relative to the RTX3090TI for the previous generation,with a die size of the RTX3060/GA106 series.
Last edited:
Caporegime
Lots of excuses makers who are emotional connected with the brand and must defend it all costs. Saw the same with when AMD overpriced Zen3 and Zen4.
Soldato
- Joined
- 30 Jul 2012
- Posts
- 2,775
So are all the dudes like myself going to call out Nvidia for their marketing bull, we gave AMD some flack, but both companies seem to be working together.
Soldato
- Joined
- 28 Oct 2011
- Posts
- 8,573
Thanks for doing the coursework I can't be arsed with.
I speculated NV planned this along, never going to release it as a 4080, appear to backtrack a little, and people just accept it's a 4070ti when it patently isn't. Sort of an anchoring tactic.
That chart has completely different 4080 die size vs other 80 die sizes than shown in today's gamer's nexus video. Time stamp 5 minutes. Its a comparison between actual size.
NVIDIA's Rip-Off - RTX 4070 Ti Review & Benchmarks
Get the LIMITED 2022 Disappointment Shirt in Cotton: https://store.gamersnexus.net/products/disappointment-pc-t-shirt-2022-100-cotton-black Or in Tri-Blend! ...
www.youtube.com
Thats because your above chart is a % vs 'titan'. Which is completely irrelevant to what we are discussing.
Caporegime
Thanks for doing the coursework I can't be arsed with.
I speculated NV planned this along, never going to release it as a 4080, appear to backtrack a little, and people just accept it's a 4070ti when it patently isn't. Sort of an anchoring tactic.
They basically took the RTX3060TI/RTX2060 Super of this generation and increased the price by over double. All the excuse makers saying But! But! 5NM costs more,don't seem to realise the GA104 Samsung 8NM was a bigger dGPU,and Samsung 8NM yields were poor too,and GDDR6/GDDR6X pricing shot up due to substrate shortages.
Because of the AD103 taking the position of the GA104 this generation,the AD104 is basically third in line like the GA106 was,and had a similar die size. Even look at the percentage of shaders/memory bandwidth vs the top card of the generation! The RTX3060TI and RTX4070TI are basically the same,and even in terms of relative performance its only really 10% here and there in the line-up.
People are trying defend what Nvidia and even AMD has been doing and now are making more absurd arguments,for something which is far worse than the Turing price hikes.
Last edited:
Soldato
- Joined
- 28 Oct 2011
- Posts
- 8,573
Yeah, plenty of Copium flying about today, nothing alters the fact it's a 4060ti for 2.5X the price of the 3060ti (which I bought). I took one look at Zen 4 prices, and decided to stay on AM4 for the next 4-5 years got 5700X and bought 32GB RAM instead, saved about £600.
The gains vs cost at the momentt are so are out of whack it's laughable.
Last edited:
Caporegime
I am on an RTX3060TI FE myself. The reason Nvidia launched the RTX4070TI as AIB models,because they are pre-overclocked out of the box. But I did some number crunching,from TPU relative performance figures.
The RTX3060TI FE was 63% of an RTX3090TI and 68% of an RTX3090 at qHD. With AIB models a few extra percent could be added.
The RTX4070TI is around 70% of an RTX4090 at qHD.
Soldato
- Joined
- 28 Oct 2011
- Posts
- 8,573
Is it though? I don't think you can seperate performance from price, like every card this gen in that regard it's a turd.
A bad card might be one with serious issues, overheating etc or terrible performance, but I don't think good performance makes for a good card without the context of price. Same with a high performance car car, or anything else really.
Thing is when looking at the current market it's hard to not spend roughly 900 quid to gain around the same performance with less effiency etc.
On the whole it's a s*** show and I'm right back to sticking it out with the 1080ti and holding out again...
On the whole it's a s*** show and I'm right back to sticking it out with the 1080ti and holding out again...
I find the arguments about $799 4070 Ti to be utterly inane, when in reality even that's not high enough as the real prices coalesce much higher. Almost as dumb as when people argue about the $699 3080s that have never existed in any real number.
Until we see the next financial report we're just flying blind vis-a-vis pricing.
Performance-wise the card's doing quite well and has reached much better numbers at 4K than I would've thought given its mem b/w, so with DLSS it will actually be a very solid card. Saw some dude with access to FG in CP2077 test it and he'd easily get that magic 120 fps (RT Ultra). It bodes very well for when that gets an official release.
More than anything the 4070 Ti is yet another card that outclasses AMD, with even the XTX not being able to keep up once RT is added, which just underscores how lost AMD is and how dominant Nvidia remains and thus they can pretty much set their prices freely. Only a very dramatic reduction in sales will really bring back some nice price cuts, but it's hard to say whether that will happen. 2023 is looking like a year of economic bloodbath so perhaps they'll relent and launch a Super line again.
Until we see the next financial report we're just flying blind vis-a-vis pricing.
Performance-wise the card's doing quite well and has reached much better numbers at 4K than I would've thought given its mem b/w, so with DLSS it will actually be a very solid card. Saw some dude with access to FG in CP2077 test it and he'd easily get that magic 120 fps (RT Ultra). It bodes very well for when that gets an official release.
More than anything the 4070 Ti is yet another card that outclasses AMD, with even the XTX not being able to keep up once RT is added, which just underscores how lost AMD is and how dominant Nvidia remains and thus they can pretty much set their prices freely. Only a very dramatic reduction in sales will really bring back some nice price cuts, but it's hard to say whether that will happen. 2023 is looking like a year of economic bloodbath so perhaps they'll relent and launch a Super line again.
Soldato
- Joined
- 28 Oct 2011
- Posts
- 8,573
I did some quick mental maths at the time and thought the 3070 wasn't worth the extra. For any gain to be "worth it" at 1440p I was going to have to go 3080/6800XT, and I wasn't doing that. Luckily without knowing much about mining the 3060ti I bought (Gigabyte) was NON LHR and got £1100 for it a year later.
Same thing now, to double 6700XT performance I'm looking at 1K with maybe £250-300 back for my card Sapphire Nitro+, nah, it doesn't add up to me. I'll look at 7900XT in a couple of years when it's going EOL.
7900 XT at ~£900 is a better buy than a RTX 4070 TI. In reality, the 7900 XT could end up being cheaper and faster.
Soldato
Anyone paying 900 quid for a 192bit interface card is just dumb to be honest..
All cards should be compared to the 6700 XT as this card represents a return to good value for money (was selling at £360 recently), and has capable all around performance.
Last edited:
Soldato
Or the the fact they use the 4080/90 massive size coolers, goes to show the huge margins on these when they strap them on. At least the Radeons come in sensible sizes.
Last edited:
- Status