NVIDIA Scrubs GeForce RTX 4080 12GB Launch; 16GB To Be Sole RTX 4080 Card
by Ryan Smith on October 14, 2022 1:20 PM EST- Posted in
- GPUs
- GeForce
- NVIDIA
- Ada Lovelace
- RTX 4080
In a short post published on NVIDIA’s website today, the company has announced that it is “unlaunching” their planned GeForce RTX 4080 12GB card. The lowest-end of the initially announce RTX 40 series cards, the RTX 4080 12GB had attracted significant criticism since it’s announcement for bifurcating the 4080 tier between two cards that didn’t even share a common GPU. Seemingly bowing to the pressure of those complaints, NVIDIA has removed the card from their RTX 40 series lineup, as well as cancelling its November launch.
NVIDIA’s brief message reads as follows:
The RTX 4080 12GB is a fantastic graphics card, but it’s not named right. Having two GPUs with the 4080 designation is confusing.
So, we’re pressing the “unlaunch” button on the 4080 12GB. The RTX 4080 16GB is amazing and on track to delight gamers everywhere on November 16th.
If the lines around the block and enthusiasm for the 4090 is any indication, the reception for the 4080 will be awesome.
NVIDIA is not providing any further details about their future plans for the AD104-based video card at this time. However given the circumstances, it’s a reasonable assumption right now that NVIDIA now intends to launch it at a later time, with a different part number.
| NVIDIA GeForce Specification Comparison | |||||
| RTX 4090 | RTX 4080 16GB | RTX 4080 12GB (Cancelled) |
|||
| CUDA Cores | 16384 | 9728 | 7680 | ||
| ROPs | 176 | 112 | 80 | ||
| Boost Clock | 2520MHz | 2505MHz | 2610MHz | ||
| Memory Clock | 21Gbps GDDR6X | 22.4Gbps GDDR6X | 21Gbps GDDR6X | ||
| Memory Bus Width | 384-bit | 256-bit | 192-bit | ||
| VRAM | 24GB | 16GB | 12GB | ||
| Single Precision Perf. | 82.6 TFLOPS | 48.7 TFLOPS | 40.1 TFLOPS | ||
| Tensor Perf. (FP16) | 330 TFLOPS | 195 TFLOPS | 160 TFLOPS | ||
| Tensor Perf. (FP8) | 660 TFLOPS | 390 TFLOPS | 321 TFLOPS | ||
| TDP | 450W | 320W | 285W | ||
| L2 Cache | 72MB | 64MB | 48MB | ||
| GPU | AD102 | AD103 | AD104 | ||
| Transistor Count | 76.3B | 45.9B | 35.8B | ||
| Architecture | Ada Lovelace | Ada Lovelace | Ada Lovelace | ||
| Manufacturing Process | TSMC 4N | TSMC 4N | TSMC 4N | ||
| Launch Date | 10/12/2022 | 11/16/2022 | Never | ||
| Launch Price | MSRP: $1599 | MSRP: $1199 | Was: $899 | ||
Taking a look at the specifications of the cards, it’s easy to see why NVIDIA’s core base of enthusiast gamers were not amused. While both RTX 4080 parts shared a common architecture, they did not share a common GPU. Or, for that matter, common performance.
The RTX 4080 12GB, as it was, would have been based on the smaller AD104 GPU, rather than the AD103 GPU used for the 16GB model. In practice, this would have caused the 12GB model to deliver only about 82% of the former’s shader/tensor throughput, and just 70% of the memory bandwidth. A sizable performance gap that NVIDIA’s own figures ahead of the launch have all but confirmed.
NVIDIA, for its part, is no stranger to overloading a product line in this fashion, with similarly-named parts delivering unequal performance and the difference denoted solely by their VRAM capacity. This was a practice that started with the GTX 1060 series, and continued with the RTX 3080 series. However, the performance gap between the RTX 4080 parts was far larger than anything NVIDIA has previously done, bringing a good deal more attention to the problems that come from having such disparate parts sharing a common product name.
Of equal criticism has been NVIDIA’s decision to sell an AD104 part as an RTX 4080 card to begin with. Traditionally in NVIDIA’s product stack, the next card below the xx80 card is some form of xx70 card. And while video card names and GPU identifiers are essentially arbitrary, NVIDIA’s early performance figures painted a picture of a card that would have performed a lot like the kind of card most people would expect from the RTX 4070 – delivering performance upwards of 20% (or more) behind the better RTX 4080, and on-par with the last-generation flagship, the RTX 3090 Ti. In other words, there has been a great deal of suspicion within the enthusiast community that NVIDIA was attempting to sell what otherwise would have been the RTX 4070 as an RTX 4080, while carrying a higher price to match.
In any case, those plans are now officially scuttled. Whatever NVIDIA has planned for their AD104-based RTX 40 series card is something only the company knows at this time. Meanwhile come November 16th when the RTX 4080 series launches, the 16GB AD103-based cards will be the only offerings available, with prices starting at $1199.
Facebook
Twitter
RSS
100 Comments
View All Comments
nandnandnand - Friday, October 14, 2022 - link
Wrong. The real scam is having a 4080 16 GB and 4080 12 GB with wildly different specs and performance. It was intended to trick people.catavalon21 - Sunday, October 16, 2022 - link
...and it's trying to trick us into being okay with a 60-series card at best passing as a 70-series even at that. 192-bit-wide memory bus on a 70 or 70ti? No thank you. For them to put it 2 tiers in the product stack above where it should be should be called what it is. We shouldn't be happy with it falling only one rung.nandnandnand - Sunday, October 16, 2022 - link
Memory bus doesn't matter, just performance. As we saw with AMD's RX 6000 series, they can skimp on the memory bus and get good performance. AMD used "Infinity Cache" (L3), and Nvidia has increased the size of their L2 cache. The RTX 3090 Ti has 6 MiB of L2 cache, and the soon-to-be-renamed 4080 12 GB card has 48 MiB.If you don't like the price/performance of a card, buy something else. The prices will fall as they sell off the now-overproduced RTX 3000 cards.
Silver5urfer - Friday, October 14, 2022 - link
Expected because 3090 and 3090Ti would offer better value and higher VRAM vs this 192Bit horrendous xx70 class AD104 chip. I'm betting the reviewers delivered their response to Nvidia. Probably Jensen realized he cannot pull of this BS scam esp in a state where the stock market cratered and Inflation is all time high.Also DLSS3 is a horrible tech, HBU did a fantastic video unlike the DF shilling campaign. The clear cut latency spike plus how the image stability already suffers with both DLSS and FSR upscalers, only worth at 4K because higher quality image and improving TAA blur BS. It won't sell the 4080 12GB because it cannot get higher FPS beyond 200 like 4090 and will suffer from smearing, image instability, image corruption, low fidelity, higher perceived latency making useless fake FPS numbers worthless.
Now the funny part is this will just get a rename as 4070Ti with probably a 50W power bump and repackaged in new cardboard box. They won't cut price unless the 40 series total stack from top get a cut because there's a hole in $1000 mark now, currently occupied by 3090Ti how long Nvidia will keep making Ampere cards ? They cannot because they want TSMC 5nn high cost dies to sell.
Another big joke is scalpers are selling these new 4090 cards at super high $2500 price and in EU the official price of 4090 itself is at 2400 Euros. Imagine buying the card at such mark and then get shafted by a full AD102 die as 4090Ti with extra $500 premium and DP2.0. Get Played by Nvidia if you support this trash company.
3090 buyers got shafted hard by experimental trash PCB design with horrendous MSVDD power rail and cancerous VRAM design making them run at 100C max yeah I know Micron rated them at 118C but nope the high temp is not valid when it doesn't have damn cooling at all. Oh the PCB also has power excursions plus Memory module placed near the PCIe slot to make it bend and eventually fail. This is how garbage Nvidia plays their consumers, yeah I'm sitting with a 3090 planning to sell it off and get a 3090Ti because I aint buying this overpriced 4080 16GB trash card and the 4090 at such high price and get shafted.
coburn_c - Saturday, October 15, 2022 - link
Now we can focus on the 70% (500 dollar) xx80 series price increase. Even if the performance is 70% better (it won't be) it's still not a value increase.nandnandnand - Saturday, October 15, 2022 - link
The price increase is to tell you to buy a 30-series card instead, unless you really want the best right now.0siris - Saturday, October 15, 2022 - link
How about cancelling the price of the other 4080 as well?yhselp - Saturday, October 15, 2022 - link
Following Ampere branding, the line-up should looks as follows:RTX 4080 12GB should be RTX 4060 (non-Ti);
RTX 4080 16GB should be RTX 4070 (again, non-Ti);
RTX 4090 should be between RTX 4080 and 4080 Ti.
Following traditional, pre-Kepler branding, the line-up should look as follows:
RTX 4080 12GB should be RTS 4050 (or RTX 4050 Ti);
RTX 4080 16GB should be RTX 4060;
RTX 4090 should be RTX 4070/80.
So really, NVIDIA needs to rebrand the other cards, too.
Adjusted for inflation and added costs, such as process, NVIDIA could still make a huge profit (more than it needs to do good business) by charging (in late 2022) the following:
$399 for RTX 4080 12GB;
$599 for RTX 4080 16GB;
$829-899 for RTX 4090.
Additionally, it might just be able to do decent business in late 2022 (after all this is the consumer, and not professional [Quadro, etc.] market) at:
$249-299 for RTX 4080 12GB;
$399 for RTX 4080 16GB;
$749-799 for RTX 4090.
michael2k - Saturday, October 15, 2022 - link
You’re suggesting a more powerful chip exists if you’re rebranding the 4090 as a 4070 or 4080.At best I think the 4090 stays unchanged, the 4080 stays unchanged, and the delaunched 4080 becomes a 4070. Then lower power parts become the 4060 and 4050.
And I don’t see why they should reprice their product, since they are selling out as is.
4090 $1600
4080 $1200
4070 $900
4060 $600
4050 $400
And of course that means the 3080, 3070, 3060, and 3050 all get slotted below them.
meacupla - Saturday, October 15, 2022 - link
Have you stopped to think that maybe, just maybe, RTX 4090 FE already draws a hefty amount of power?It's already par with a 3090Ti at this point, with maximum 500W power consumption, and that's before any factory overclock is applied.
Like, realistically, what do you think a "4090 Ti" with your lineup will be?
A 650~750W card that requires 2x 12VHPWR?