Nvidia’s RTX 5070 is seemingly going to be revealed at CES 2025, alongside the rumored RTX 5080 and 5090 next-gen graphics playing cards.
This can be a notable change in the way in which the rumor winds are blowing, as up till now, we’ve been listening to that it’s simply the RTX 5080 and 5090 that’ll be the preliminary Blackwell GPUs to pitch up from Nvidia.
Wccftech has made this declare, and in addition supplies some spec revelations for the RTX 5070, after chatting with sources (who yesterday delivered a leak on the RTX 5080 getting speedier VRAM). Moreover, two leakers on X have chimed in, not disagreeing with the tech web site, and offering their very own tackle the spec of the RTX 5070.
Wccftech believes that the RTX 5070 will use the GB205 chip, however doesn’t give us a core depend. We’re additionally informed that the graphics card will run with a 192-bit reminiscence bus and 12GB of VRAM (GDDR7). This would be the slower taste of GDDR7, 28Gbps, which can even be on board the RTX 5090 (however the 5080 will get the 32Gbps reminiscence, if Wccftech’s sources are right, that’s).
For the RTX 5070, then, we’re a complete reminiscence bandwidth of 672GB/s, fairly a bit lower than the RTX 5080’s supposed 1024GB/s, however a very good chunk greater than the RTX 4070 (a 3rd extra, in actual fact).
Nevertheless, we’re informed energy utilization can even creep up for the RTX 5070, and that it’ll chug 250W (in comparison with 200W for the RTX 4070).
Leaker Harukaze5719 backs up these specs fully, whereas including a purported core depend, specifically 6,400. One other common leaker on X, Kopite7kimi, backs up the 250W TDP idea, too (and the 12GB of VRAM assertion, for that matter).
GB202: 24,576 RTX 5090: 21,760 / GB202-300-A1 / 512-bit / 32GB GDDR7, 28Gbps, ~1.8TB/s / 600W TBP / 14-Layer FE PCBGB203: 10,752 RTX 5080: 10,752 / GB203-400-A1 / 256-bit / 16GB GDDR7, 32Gbps, 1TB/s, 400W TBPGB205: 6,400 RTX 5070: TBD / GB205 Primarily based / 192-bit / 12GB GDDR7,… https://t.co/KTO6QLdKCkOctober 9, 2024
As to the arrival timeframe, do not forget that that is referring to the preliminary unveiling of those three Blackwell graphics playing cards collectively at CES 2025 (in idea). In terms of the on-sale dates, the RTX 5090 and 5080 are supposedly going to be shut collectively, however we’d think about the RTX 5070 is more likely to be additional away given its very sudden look within the Blackwell launch rumors.
Evaluation: Nvidia’s RTX 5070 – a sheep in wolf’s clothes GPU?
Okay, deep breaths. As you may think about, there’s been some displeasure at these RTX 5070 specs, and notably with the reminiscence (and the drop to the GB205 chip – the RTX 5080 will use GB203).
You may guess the sort of stuff: ‘Nvidia is scamming PC players once more!’ ‘The RTX 5070 ought to have 16GB of VRAM, the RTX 5060 ought to have 12GB – that is an RTX 5060 being offered as a 5070!’ And so forth…
Properly, these individuals have a degree, definitely – however we are able to’t get forward of ourselves right here. We’re nonetheless in rumor-land, firstly, and in addition, we don’t know the pricing connected to those GPUs. Regardless of the respective specs, Nvidia will hopefully be certain that these graphics playing cards make sense when it comes to efficiency per greenback, and total worth.
Though we guess the difficulty is that almost all people discover it onerous to imagine Nvidia will strike an inexpensive worth proposition, and that features us. This might imply a elementary change in angle for Group Inexperienced, and there must be a purpose for that. The one chance is RDNA 4’s efficiency and worth stability – however bear in mind, AMD will prime out on the mid-range. That means it may need some affect on the RTX 5070, maybe, however for the higher-end Nvidia Blackwell playing cards, there received’t be any rivals, not remotely.
RelatedPosts
So, in brief, we absolutely perceive the concerns right here, however let’s not condemn these graphics playing cards primarily based on rumors earlier than they even arrive, nevertheless tempting that may be.
Nonetheless, whether it is true that the RTX 5070 can have 12GB, and RTX 5080 16GB, you’ve received to ask your self: why? This actually doesn’t really feel like sufficient VRAM at this time limit for PC players, not to mention a remotely future-proofed loadout because the years roll on (and we’ll have a very good whereas till RTX 6000, after all).
The cynical will, after all, counsel that these video RAM configurations give Nvidia room to maneuver down the road with ‘Tremendous’ or ‘Ti’ variations. There may be rumored to be a 24GB variant of the RTX 5080 coming later, which may be the RTX 5080 Tremendous, and this permits Nvidia to cost extra for what most players would see as the correct quantity of VRAM for this degree of graphics card within the present gaming local weather. Maybe it’ll be ditto for a 5070 packing 16GB of video reminiscence, all of which looks like cynical techniques to maximise gross sales and income.
This raises one other specter – after we do get the RTX 5060 arriving, presumably that’s going to have 8GB, then? As if there weren’t sufficient complaints about sticking 8GB on the present RTX 4060. Yeah, we’re sort of getting sucked into the unfavorable spiral of worries right here, however we should preserve reminding ourselves – these are simply rumors. And Nvidia should guarantee these graphics playing cards make some sort of sense, when it comes to their total efficiency and price ticket.
Nevertheless, it does really feel like Nvidia is pushing players who play at 4K to purchase their priciest GPUs, and you’ll wager that these top-end Blackwell choices are going to be significantly costly.
On a last, extra constructive observe, Wccftech talks about Nvidia revealing “next-generation AI applied sciences” for the PC gamer which shall be an enormous shock at CES 2025. Colour us intrigued. A manner for DLSS 4 to make up for lean VRAM loadouts, maybe?
By way of VideoCardz
You may additionally like
