This month, Nvidia will release three new graphics models of the Ada Lovelace generation – refreshed graphics labeled Super. However, this autumn it will be two years since the 4000 generation came to the market, and it will be time to look for a successor. Youtuber Moore’s Law is Dead has now come up with information about when this new generation can be expected on the market. It looks like it could be up to the competing RDNA 4.
Nvidia has recently always released a new generation exactly after two years (since the RTX 2000 it’s always at the turn of summer and autumn), but this year it might not be followed. Earlier there was information that this time the new generation will be later, in 2025. It is not just some gossip, because this was directly indicated by Nvidia’s official roadmap, so it is the company itself that started these discussions about the relatively later arrival of the new generation.
According to Moore’s Law is Dead, it is true that Ada Lovelace is scheduled for release in 2025, so the normal 24-month cycle has been extended and it was not just a mistake. But the details are important.
Nvidia product roadmap. The GeForce RTX 5000 is labeled Ada Lovelace-Next
Author: Nvidia, via: Tom’s Hardware
Sources from Nvidia reportedly told Moore’s Law is Dead that Nvidia is likely to announce a new generation of graphics cards at CES 2025, the January trade show in Las Vegas next year. This does not necessarily have to be the release date, the cards can realistically be on sale in a few weeks, or even in a month. According to the source, however, Nvidia will want to loudly “present the energy efficiency of this new generation” at CES. If this is true, it makes sense to do it only when the graphics can already be bought or are almost on the shelves, otherwise it would rather devalue the previous generation that is still being sold.
It is said that the cards should be released sometime in early 2025, but internally the development is not going so that this is the earliest possible release date. On the contrary, it is said that Nvidia is working with a margin so that, if necessary, the release can be moved to the last quarter of this year, 2024, although it is not clear whether this means December or perhaps already October (in which case, in the end, there would be no lag behind the traditional did not occur during the two-year cycle).
According to various rumors, AMD could release its own new generation RDNA 4 (Radeon RX 8000) already in the second half, or rather in the fourth quarter of 2024. Moore’s Law is Dead says that this very event could theoretically force Nvidia to speed up the release bid for the current year.
However, currently the RDNA 4 is rumored to contain only two chips Navi 44 and Navi 48 intended for, say, the lower half of the market. According to leaker Kepler, these chips are said to be similar in size to the Navi 23 (237 mm² on 7nm process) and Navi 24 (107 mm² on 6nm process) chips from the RDNA 2 generation, and both are monolithic. Therefore, they will not compete with the GeForce RTX 5000 graphics, which Nvidia, as usual, will probably start introducing from high-end models with multiple larger GPUs downwards.
Less progress than RTX 4000?
Moore’s Law is Dead also gives some shots at where Nvidia is said to be headed with performance. The increase in performance with the new generation GeForce RTX 5000 could probably be smaller than between Ampere and Ada Lovelace, which is supposed to be an indication of rough performance in raster graphics, i.e. without including any new tricks like DLSS or ray tracing improvements. This probably refers to a theoretical comparison in which the high-end GPU of both series (AD102 and GB202) would have all units active.
GeForce RTX 4090
Author: Nvidia
Here, however, it is important that the current generation of graphics cards does not make much use of the AD102’s potential, the GeForce RTX 4090 has only 128 active out of 144 chip units. Nvidia is said to be able to do this so that in the GeForce RTX 5000 generation, the top model will not be artificially held back in this way. And that would then increase the intergenerational increase between Ada Lovelace (RTX 4090) and Blackwell (hypothetical GeForce RTX 5090).
But beware that this is youtube information coming a year before the release. The specifications are certainly not final yet and may change. Certainty of how performance will turn out is now probably minimal, and current internal performance predictions are also likely to be conservative. That’s why we think you basically don’t need to remember this performance estimate at all, it’s probably not very relevant even if it’s a legitimate number straight from Jen-Hsun Huang. At the moment, the performance is simply not determined with such great precision, it will be more of a target (which is not completely binding, see IPC increases for AMD Zen cores), if at all.
We recently discussed in a recent article what parameters the leading most powerful graphics of the GeForce RTX 5000 generation could have:
Tip: New information on the GeForce RTX 5090: Blackwell chips are 3nm, but they don’t have 512-bit memory after all
Even here, of course, things can still change, although these parameters come from the leaker Kopite7kimi, which is basically the best source of internal information from the Nvidia circuit.
Resources: TechPowerUp, Moore’s Law is Dead, Kepler_L2
2024-01-03 05:06:13
#GeForce #RTX #year #CES #Nvidia #tactical #release #year #Cnews.cz
