• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Nvidia RTX 6000 series (codename Rubin)

So given this whole AI thing and shortage of components and escalating prices a 2026 release is probably not on the cards right.

So maybe CES 2027?
 
Last edited:
opethdisciple said:
So given this whole AI thing and shortage of components and escalating prices a 2026 release is probably not on the cards right.

So maybe CES 2027?
Click to expand...

If they're launching late 2027 like even the well established leakers are claiming then I'm guessing we'll hear solid stuff about them at Nvidia's GTC around March 2027.
 
Media reports Nvidia has signed a deal with Intel to partially build its 2028 Feynman architecture GPU's on its 14A node

To avoid confusion, Feynman GPUs would be RTX7000 but we don't yet know have a RTX7000 thread

www.digitimes.com

Exclusive: Nvidia to reportedly shift 2028 chip production to Intel, reshaping TSMC strategy

TSMC's dominance in advanced process and packaging has made it a prime target amid US manufacturing mandates. Chip customers now face mounting pressure to diversify supply chains due to cost and capacity constraints, accelerating the shift toward multi-sourcing strategies.
www.digitimes.com
 
Grim5 said:
Media reports Nvidia has signed a deal with Intel to partially build its 2028 Feynman architecture GPU's on its 14A node

To avoid confusion, Feynman GPUs would be RTX7000 but we don't yet know have a RTX7000 thread

www.digitimes.com

Exclusive: Nvidia to reportedly shift 2028 chip production to Intel, reshaping TSMC strategy

TSMC's dominance in advanced process and packaging has made it a prime target amid US manufacturing mandates. Chip customers now face mounting pressure to diversify supply chains due to cost and capacity constraints, accelerating the shift toward multi-sourcing strategies.
www.digitimes.com
Click to expand...
by the time that is released there will be no PC anymore, ram will likely be £2000 for 32GB and RTX7000 will start at £7000 and we will all be on subscription service renting PC from the cloud.
 
Last edited:
Not sure if this has cropped up in other threads, but seen that some 6 series have been speculatively added to TPU database recently... has there been an announcment that I've missed?

www.techpowerup.com

NVIDIA GeForce RTX 6070 Specs

NVIDIA GR205, 2600 MHz, 6144 Cores, 192 TMUs, 80 ROPs, 16384 MB GDDR7, 1750 MHz, 256 bit
www.techpowerup.com

www.techpowerup.com

NVIDIA GeForce RTX 6080 Specs

NVIDIA GR203, 2800 MHz, 10752 Cores, 336 TMUs, 112 ROPs, 20480 MB GDDR7, 1875 MHz, 320 bit
www.techpowerup.com

www.techpowerup.com

NVIDIA GeForce RTX 6090 Specs

NVIDIA GR202, 2700 MHz, 21760 Cores, 680 TMUs, 176 ROPs, 32768 MB GDDR7, 1750 MHz, 512 bit
www.techpowerup.com


Edit: Just seen that core counts are identical to 5 series, just a bump on memory bus size for the 70 and 80.. guess it's a nothing burger then!
 
Last edited:
No news and we've had no leaks about them either

You can kind of guess core count at least using the specs NVIDIA release for Rubin data centre parts but you also have to make an assumption about whether rtx6000 is monolithic or MCM as some Rubin data centre cards are MCM
 
Last edited:
Grim5 said:
No news and we've had no leaks about them either

You can kind of guess core count at least using the specs NVIDIA release for Rubin data centre parts but you also have to make an assumption about whether rtx6000 is monolithic or MCM as some Rubin data centre cards are MCM
Click to expand...
Yeah, per my edit, it's clearly just a copypasta from 5 series specs with a dash of optimism on the memory bus... Just confused about why TPU have added them.

Hoping they can nail down MCM with enough bandwidth for GPUs, hopefully bring us back to the fun times when they could just slap two cores on a PCB and release something mental and at a better value than the top-end die! (i.e. 7950GX2, Titan Z etc.)
Been hoping it'd go that way since Ryzen first came out, then there were the rumours AMD would deliver on the 5/6k series... as far as I understand it, it's the interconnect that's the bottleneck.. must be a reason SLI/CF hasn't resurfaced!
 
TrojanWhore said:
Just confused about why TPU have added them.
Click to expand...
Just clickbait. No one really knows what the specs are as always and the usual made up rubbish.

The whole industry is in self destruction and only way these sites and youtube channels are getting clicks.

Till leather jacket man comes out and says what's what and more fake frames and more planned obsolescense features that only work on new GPUs, then we will know.
 
Last edited:
Thing is Nvidia and AMD will prototype many different designs to test the architecture and manufacturing processes so even IF the social media news has a shred of truth to it (and I'm being very generous here most of it is hedging opinionated speculation) its of zero value up until they commit to mass manufacturing. That is what you really need to know followed by the price which again isn't advertised until weeks before release and even then is shrouded in bogus marketing smoke & mirrors due to supply constraints.
 
You must log in or register to reply here.
Back
Top Bottom