NVidia Ada Speculation, Rumours and Discussion

Status
Not open for further replies.
Gamescom launch it is, then, if that bit of information is to be trusted, of course.
Not even VideoCardz believe in it, though. Also I wonder what chinese supplier is willing to ship anything to Russia, get caught and you might not get cards at all anymore (all major tech companies have forbidden shipping their products there).
 
Not even VideoCardz believe in it, though. Also I wonder what chinese supplier is willing to ship anything to Russia, get caught and you might not get cards at all anymore (all major tech companies have forbidden shipping their products there).
They will be shipped woth intermediaries, of course. Maybe the card was "sold" in (insert name of a country served by Chinese suppliers not sanctioning Russia here) and then delivered to the real buyier in St. Petersbourg or Moscow. Nothing new under the sun.
 
Not even VideoCardz believe in it, though. Also I wonder what chinese supplier is willing to ship anything to Russia, get caught and you might not get cards at all anymore (all major tech companies have forbidden shipping their products there).
I don't think this is true actually but I haven't followed recent advancements - these things tend to change over night.
AFAIR it was basically Nvidia who stopped shipping and selling their cards in Russia (FE models).
Most AIBs haven't said anything on the matter and there are no restrictions imposed on selling GPUs to end users, only to country owned entities.
 
Pretty sure that's software stuff, not actually related to Ada hardware
 
I don't know how you can rule out either way with that information. Nvidia did first announce Turing, but not RTX 2xxx which came slightly later at Gamescom, at SIGGRAPH 2018 with a similarly on paper unassuming lead off address.

Turing looking back was a very a professional and prosumer oriented launch with then Quadro cards actually coming first as opposed to months or longer after consumer Geforce cards. It's also been argued that Nvidia's inclusion of ray trace and machine learning functionality was of greater benefit at that stage to pro/prosumer use than gaming while the more traditional gaming perf (non RT) improvements nad perf/$ were relatively poorly received.

Given the current market context this also has additional implications as doing it with this focus would effectively mean they can maintain schedule (as they've already confimred they will transition architectures this year) and launch Ada with a professional and prosumer focus while still having Ampere mainly for consumers during the Q4 holiday season.
 
Last edited:


Just wait and see how the new gpu's will perform, you will see a large IHV like nv can do.
 
Seems the reason behind they sticking with PCIE Gen 4 in Ada is to reuse RTX 3000's board designs. Sounds reasonable if it could reduce cost increase profits and get faster release cycle, while it doesn't hurt performance at all.
FTFY
 

You gotta love the "configuration changes" when there is literally no configuration to change yet.
Also such change doesn't mean anything for TDP apparently.
 

You gotta love the "configuration changes" when there is literally no configuration to change yet.
Also such change doesn't mean anything for TDP apparently.

Amazing right. Which is more likely? Nvidia changing their minds every 2 days. Or “leakers” being clueless attention whores.
 
Status
Not open for further replies.
Back
Top