News Most RTX 50-series GPUs sold out in five minutes at Newegg — entire inventory evaporated in just 20 minutes

[Admin]

Newegg's RTX 50 stock was gone in 20 minutes, raising questions about the quantity available at launch.

Most RTX 50-series GPUs sold out in five minutes at Newegg — entire inventory evaporated in just 20 minutes : Read more

 

[Co BIY]

They are probably underpriced at MSRP. (ducks in anticipation)

 

[DS426]

It's already known at this time that 5080 and 5090 are a paper launch.

 

[Jabberwocky79]

Welp, I was hoping to pick one up in March.... but even that might be optimistic.

 

[why_wolf]

Feels like Nvidia was desperate to both get a product out to keep the stock price high but also to get in before any new tariffs are levied. To establish a facts on the ground position. Its harder to slap a tariff on something that is already for sale as opposed to something that is not yet for sale. Because the customer will notice the obvious price increase and get mad.

That coupled with the Chinese being on their annual month long holiday means none of the factories are capable of working at full capacity.

 

[OneMoreUser]

I hope no one buys from the scalpers, so that instead they end taking a loss.

 

[Jabberwocky79]

I hope no one buys from the scalpers, so that instead they end taking a loss.

Well, there's a few people out there fighting the good fight... there are several listings that are similar to this one: https://www.ebay.com/itm/1768059520...68ikFSuSDSx56CITsRhozSkg==|tkp:Bk9SR5qnlueXZQ

Unfortunately however, the first 120 sold listings of 5090s on eBay reveal a top price of $9000 and a bottom price of $3800. There have been 269 sales so far, but some of them are bogus listings like the one I linked to.

 

[beyondlogic]

ah dont worry ive already seen 5080s going for £1,399.99 on ccl in uk bloody attrocious lol.

Palit GeForce RTX 5080 GameRock 16GB OC GPU - NE75080T19T2-GB2030G

Available @ CCL, this Palit GeForce RTX 5080 GameRock 16GB GDDR7 Overclocked Graphics Card supports Raytracing, is VR Ready and has HDMI and outputs, Mfg Code: NE75080T19T2-GB2030G

www.cclonline.com

Zotac GeForce RTX 5090 Solid 32GB GDDR7 PCI-Express Graphics Card​

GRA-ZTC-04049
£2,239.99

https://www.overclockers.co.uk/pc-c...eforce-rtx-5090-graphics-cards?sort=price_asc

 

[acadia11]

I don’t know about anyone else but I never saw any stock listed on any of these sites, they all said sold out the minute you looked. if Intel had any since as a foundry they would have been priming a long time ago to support the likes of nvidia. What I don’t understand is why Nvidia is exclusively going to TSMC for foundry services , it’s got to be the Taiwan good old boys network. While there is a global chip demand crunch across the board … it seems to be in part due to contracts and loyalties.

 

[Jabberwocky79]

it seems to be in part due to contracts and loyalties.

Yep, that's the way just about every form of business works. I've seen my superiors make some really boneheaded decisions over the years due to an unhealthy loyalty to a certain company or sales rep.

 

[beyondlogic]

I don’t know about anyone else but I never saw any stock listed on any of these sites, they all said sold out the minute you looked. if Intel had any since as a foundry they would have been priming a long time ago to support the likes of nvidia. What I don’t understand is why Nvidia is exclusively going to TSMC for foundry services , it’s got to be the Taiwan good old boys network. While there is a global chip demand crunch across the board … it seems to be in part due to contracts and loyalties.

tsmc is much larger.

and its not like they havent thought of using other companys i know they have had talks with samsung

 

[fiyz]

When you control the demand, creating scarcity only makes the product that much more in demand. He can charge Elon Musk and all those AI companies a premium for the latest and greatest. If he doesn't milk them, who will?

And to: "What I don’t understand is why Nvidia is exclusively going to TSMC for foundry services" by Acadia,

Tsmc has some of the best yields from their silicon. They are way ahead of intel, though intel is improving. Still, their yields arent as impressive outside of Taiwan, and more or less on par with Intel. You gotta realize, they're firing lasers at motes of dust to slam them into the substrate, with enough accuracy to build multilayered circuits.

 

[rluker5]

You know what else that ends in egg that is in short supply and overpriced?
Eggs.
A little over an hour ago I picked some up and since they were price/pound competitive I went with duck eggs.
Hopefully they don't suck.

Also my hard to find B580 is a bunch of fun. I moved it from an MSI to an Asus motherboard and my pc's sleep related issues with that card went away. When OCd is almost as good as my 3080 in CP2077, but underperforms my A750 in Guardians of the Galaxy. I'm sure that second game will get fixed. Been running it upscaled 4k60 with highish settings.

 

[Trake_17]

YOU'RE MISSING OUT! YOU MUST DESPERATELY SEEK ONE AT ANY COST!

 

[Gaidax]

I don’t know about anyone else but I never saw any stock listed on any of these sites, they all said sold out the minute you looked. if Intel had any since as a foundry they would have been priming a long time ago to support the likes of nvidia. What I don’t understand is why Nvidia is exclusively going to TSMC for foundry services , it’s got to be the Taiwan good old boys network. While there is a global chip demand crunch across the board … it seems to be in part due to contracts and loyalties.

It's really simple.

Because TSMC is better and more reliable.

And more specifically - TSMC's 4N/4NP process is considered mature with proven yields, while Intel's comparable Intel 4 process lags behind, is not even commercially available for anyone but Intel itself and no track record outside of Intel.

On top of that TSMC already has 3nm locked and commercially available with mass production for a year. Intel is not even near that for now.

 

[acadia11]

tsmc is much larger.

and its not like they havent thought of using other companys i know they have had talks with samsung

Exactly Samsung has the largest leading edge foundary capacity in the world … so what does that tell you? They were in talks with Intel as well and Samsung … and are now exclusively TSMC

 

[acadia11]

It's really simple.

Because TSMC is better and more reliable.

And more specifically - TSMC's 4N/4NP process is considered mature with proven yields, while Intel's comparable Intel 4 process lags behind, is not even commercially available for anyone but Intel itself and no track record outside of Intel.

It's really simple.

Because TSMC is better and more reliable.

And more specifically - TSMC's 4N/4NP process is considered mature with proven yields, while Intel's comparable Intel 4 process lags behind, is not even commercially available for anyone but Intel itself and no track record outside of Intel.

It's really simple.

Because TSMC is better and more reliable.

And more specifically - TSMC's 4N/4NP process is considered mature with proven yields, while Intel's comparable Intel 4 process lags behind, is not even commercially available for anyone but Intel itself and no track record outside of Intel.

On top of that TSMC already has 3nm locked and commercially available with mass production for a year. Intel is not even near that for now.

It’s a general comment not specific to 4n or intels planned 18A , it’s Monday morning quarterbacking Nvidia can push the industry on the direction it wants not just limiting that influence to TSMC. Samsung is as strong and Intel foundry was for quite sometime producing some of the most advanced leading edge process nodes.

 

[Gaidax]

Exactly Samsung has the largest leading edge foundary capacity in the world … so what does that tell you? They were in talks with Intel as well and Samsung … and are now exclusively TSMC

There are actually a few articles over the past week where it's revealed that Jensen is unhappy with Samsung performance both on memory front and foundry as well, and in a pretty colorful language at that.

It’s a general comment not specific to 4n or intels planned 18A , it’s Monday morning quarterbacking Nvidia can push the industry on the direction it wants not just limiting that influence to TSMC. Samsung is as strong and Intel foundry was for quite sometime producing some of the most advanced leading edge process nodes.

It a simple business. What was mass produced with Intel 4 so far? Mobile 14th gen Intel CPUs and that's it.

Meanwhile TSMCs comparable N4 was used to produce all the Series 40 GPUs, all Apple's Bionic 16 from 2 years ago (so all iPhones, iPads and so on) AND all the Snapdragon 8+ Gen1, so practically a good chunk of all smartphone SoCs in existence.

And it's not even the highest end commercially available process for TSMC, while Intel 4 is the best they got that even seen the light of day in some sort of production.

People think it's just some feels or favorites here. No, it's a hard cold reality of TSMC having commercial availability AND insane track record of delivering as shown above.

 

[acadia11]

There are actually a few articles over the past week where it's revealed that Jensen is unhappy with Samsung performance both on memory front and foundry as well, and in a pretty colorful language at that.

Hearsay, I’m going to go with good old boys network theory. But that’s not point really … it’s an observation that nvidia could if it wanted source from multiple vendors, although not sure that would mean from a design standpoint as I understand there is tight coupling between the production features TSMC process provides and the actual end product … so who knows.

 

[Gaidax]

Hearsay, I’m going to go with good old boys network theory. But that’s not point really … it’s an observation that nvidia could if it wanted source from multiple vendors, although not sure that would mean from a design standpoint as I understand there is tight coupling between the production features TSMC process provides and the actual end product … so who knows.

I mean, I literally wrote above why TSMC is on the next level.

How exactly they "could have" done that, if Intel does not even have anything comparable to N4P even available for orders outside of Intel? Let alone no real track record for it.

It's a simple fact that TSMC is a good year or two ahead of their competitors and that's why they farm all the important orders.

 

[YSCCC]

It’s a general comment not specific to 4n or intels planned 18A , it’s Monday morning quarterbacking Nvidia can push the industry on the direction it wants not just limiting that influence to TSMC. Samsung is as strong and Intel foundry was for quite sometime producing some of the most advanced leading edge process nodes.

Actually I believe those are more contractual issues on their side, it's common to be exclusive to one, and if I am Nvidia I will do so also to secure a better deal, and more importantly, ppl won't go look for ridiculous stuffs like : I want the TSMC 5090, not Samsung 5090 or let ungrounded rumours of which have better OC capability etc. circle around.

But then using whom as a foundary is less of an issue on the street pricing, they started that scalping war and makes scalpers like shark smelling blood, it takes a gen or two with no realistic scalper market just like how iphones do back in the days to stop this overpriced crap. (we have scalpers literally go bankrupt and begs to sell a a loss with piles of iphones on the street, I forgot was it iphone 7 or 8..). All we could do is not paying a car worth of money for a gaming GPU to begin with

 

[Elusive Ruse]

40 units sold in 20 minutes, that's crazy! The Nvidia RTX 50 series has to be an incredible offering me thinks!

 

[rluker5]

There are actually a few articles over the past week where it's revealed that Jensen is unhappy with Samsung performance both on memory front and foundry as well, and in a pretty colorful language at that.


It a simple business. What was mass produced with Intel 4 so far? Mobile 14th gen Intel CPUs and that's it.

Meanwhile TSMCs comparable N4 was used to produce all the Series 40 GPUs, all Apple's Bionic 16 from 2 years ago (so all iPhones, iPads and so on) AND all the Snapdragon 8+ Gen1, so practically a good chunk of all smartphone SoCs in existence.

And it's not even the highest end commercially available process for TSMC, while Intel 4 is the best they got that even seen the light of day in some sort of production.

People think it's just some feels or favorites here. No, it's a hard cold reality of TSMC having commercial availability AND insane track record of delivering as shown above.

Intel 3 says hi from June last year:
https://www.tomshardware.com/tech-i...technology-is-in-high-volume-production-intel

 

[Gaidax]

Intel 3 says hi from June last year:
https://www.tomshardware.com/tech-i...technology-is-in-high-volume-production-intel

What chips Intel 3 has mass produced and are commercially available?

It's kind of sad; people aren't getting it - what is needed here is a track record, not empty statements. That's the difference between TSMC and Intel on this front.

 

[ezst036]

Hearsay, I’m going to go with good old boys network theory.

Hearsay, I'm going to go with hearsay.

I'm not saying the other comment is correct. I'm saying you exchanged hearsay you didn't like for hearsay you preferred. "My hearsay is better than your hearsay." <-- This is exactly what you said.