Nvidia’s RTX 40 series is expected to be launched in July

The rumor mill is churning as the launch of the next-gen RTX 40-series graphics cards based on the Ada Lovelace architecture approaches. New information suggests the next-gen cards will launch even sooner than expected, perhaps August if not July.

The new rumor comes courtesy of the well-known leaker Kopite7kimi, who mentioned the early third quarter, which could indicate a July start. A Gamescom launch in August is also a clear possibility, but that would mean a mid-to-late Q3 launch. There’s always Jensen’s kitchen. Assuming the launch occurs in either early or mid-Q3, it’s coming sooner than we anticipated, with September or October previously considered a likely launch window. Launching earlier would almost certainly give it a huge advantage over competitor AMD, and possibly even beat Intel in launching its Arc series of cards.

Let’s not forget that Computex returns this year, albeit in a reduced format due to the ongoing Covid considerations. Nvidia is expected to deliver a keynote speech on May 23rd. One of the speakers is Jeff Fisher, Senior Vice President of GeForce. We’re unlikely to get any RTX 40 information, though, as Jensen himself is the one likely to announce it. Don’t rule out the possibility of a teaser, though.

It is believed that the high-end cards will come out first. They make Halo headlines and the production numbers don’t have to be as high as for a mainstream launch. This is backed up by the fact that we’ve heard little about the other GPUs in the line-up beyond a few codenames and possible shader counts.

All signs point to the high-end AD102 being a GPU monster. It used to be speculated that Ada Lovelace was closer to an evolution of Ampere than something entirely new, but that seems to have changed in recent months. Nvidia seems to be doing everything it can to retain its fastest GPU crown. Sure, a 900W GPU is a step too far, but now that the RTX 3090 Ti has surpassed the 450W threshold, those 600W rumors are becoming more and more true.

A TDP of 600W would mean that third-party cards might include dual 16-pin power connectors for well over 1kW of possible power output. Just let that sink in. The RTX 4090 may need any of those 600 watts to power it. Up to 18,432 cores and 96MB L2 cache alone would provide dramatic performance gains over the 450W RTX 3090 Ti’s 10,752 cores and 6MB L2 cache. But add other architectural improvements, revamped RT and Tensor cores, and there’s no doubt that the RTX 4090 will be a pixel-pushing monster of a card. AMD will have their hands full trying to take the throne.

This article was previously published on Source link