Nvidia’s RTX 4080 is allegedly getting a new spin on the GPU which could bring the cost down — but any price cut would likely be very slight, unfortunately, if it happens at all.
tom devices (Opens in a new tab) This rumor has been reported—and treated with caution, as with anything from a continuous spinning mill—that it originated from HKEPC (Opens in a new tab) (Hong Kong tech site), claiming that while the current RTX 4080 graphics card is built on the AD103-300 chip, Nvidia will use a slightly different GPU in the future, namely the AD103-301.
There’s now more evidence that this is actually happening, Tom points out: Graphics card maker Galax has a graphics processing unit (GPU) named “AD103-300/301” under its RTX 4080 product detail listings.
Furthermore, VideoCardz (Opens in a new tab)which also picked up on this, tells us that another card manufacturer Gainward has also included its updated AD103-301 GPU variant in the product specifications.
With a couple of separate third-party graphics card makers mentioning this new spin on the GPU in their specs, it seems very likely that this is indeed happening – but of course it still has to be confirmed by Nvidia.
HKEPC makes another claim that the incoming RTX 4070 (the vanilla version of the RTX 4070 Ti, which was just revealed) may also take the approach of having two different chipsets, which in this case would theoretically be the AD104-250 and AD104-251. The difference is that it’s supposed to The former uses a comparator circuit, but the latter does not, and will use a different circuit board, which may reduce the cost of making the card.
The same goes for the new RTX 4080 and AD103-301, with the potential to bring down the cost of building it (the bill of materials or bill of materials as it’s known in the industry). The main question is, if all of this works for either or any of the GPUs, what price could the second variants actually be?
Analysis: Does this mean cheaper graphics cards – or is this just a pipe dream?
HKEPC believes that the cost gap between the supposedly different versions of the RTX 4070 could be less than a dollar in terms of bill of materials, which is clearly a small amount, and likely makes no difference to the price. The consumer pays. A source used by VideoCardz also claims that any difference in cost wouldn’t be huge.
We haven’t given any figures for the potential difference in how much it costs to make versions of the RTX 4080, but the new AD103-301 could lower the bill considerably, and possibly mean the graphics card gets even cheaper for the consumer. However, this could be a similarly very slight downgrade in the item list of components, which no doubt won’t affect much the price of the second spin on the RTX 4080.
Time will tell, but given Nvidia’s stance on pricing with Lovelace GPUs in general, it’s hard to be optimistic at all.
Another question given to change to the RTX 4080 chip, will it affect the performance? The rumor mill is very clear in its assertion that it won’t, nor will the AD103-301 chip be more performant than the current AD103-300 chip. It makes sense, as Nvidia is supposed to be wary of dropping a dose of buyer’s remorse on those who’ve already forked out for the RTX 4080, and splashed a big wedge of cash to do so.
The overall result of this is likely to be graphics cards with the same level of performance—or so similar as to be negligible—and slight savings for board makers, which likely won’t be passed on to consumers. Hopefully, it’s more convenient, despite the cost, so keep your fingers crossed if you’re in the market for a higher-end GPU.