The “Order of 10”

[FatherMurphy]

We were told GDDR5X enters mass production in the summer. We're still in spring.

Either the timetable moved ahead significantly and no one told us, or NVidia and partners don't have chips to put on the boards yet, or at least don't have them in meaningful numbers.

Don't have time to pull the quote, but Micron announced mass production/revenue by the end of their fiscal quarter ending early May, I believe. So, it appears GDDR5X will be available in volume shortly. If Nvidia launches Pascal with GDDR5X at Computex, as speculated, then it probably is going to use some risk production samples. There will probably be shortages of any chip with GDDR5X for a month or two after that.

The speculation is that only the 1080 will use GDDR5X and the 1070 will use vanilla GDDR5. The 1080 (knowing Nvidia) will be priced with a premium, resulting in it being a lower volume card than 1070.

I think the "Order of 10" thing is neat, but if Nvidia is suggesting that Pascal is 10x faster than Maxwell, I'm skeptical to say the least. Curious to see what weird metric they use (like the "CEO Math" JH referenced last year) to arrive at that figure, if that's the case.

 

[RaulF]

We were told GDDR5X enters mass production in the summer. We're still in spring.

Either the timetable moved ahead significantly and no one told us, or NVidia and partners don't have chips to put on the boards yet, or at least don't have them in meaningful numbers.



The 200 series launched Dec 11 with retail availability Jan 12. Fury launched and no one could buy it due to HBM1 shortages for at least two months. That one feels familiar to the current memory situation.

I remember being able to buy Fury. But yeah you are right no one could buy it.

 

[Despoiler]

With Polaris announced/demoed in December, we already know :biggrin:

No one considers a demo the same as the launch announcement.

 

[S.H.O.D.A.N.]

I remember being able to buy Fury.

Good for you. Doesn't change the fact the card was hard to come by for quite a while due to yield issues on HBM1.

 

[Jaydip]

So weren't we led to believe that Polaris was six months ahead of NV? it seems everyone got that wrong, NV knows to keep their secrets

 

[airfathaaaaa]

Good for you. Doesn't change the fact the card was hard to come for quite a while due to yield issues on HBM1.

even radeon pro duo is reporting shortages on some sites because of hbm again (that or they really didnt expected much )

 

[antihelten]

That's it? All a person had to do to get there was go to www.orderof10.com even before the puzzle was solved. I was expecting something more in the end.

Right now you can of course get to the countdown page straight from the orderof10 site, but originally this wasn't possible.

Full story here:
http://www.gamersnexus.net/industry/2419-mystery-orderof10-box-mentions-fermi-we-are-decoding-it

 

[S.H.O.D.A.N.]

So weren't we led to believe that Polaris was six months ahead of NV?

I don't know a single reputable source that has claimed that. Clickbait sites and forum warriors don't count.

In fact, the whole 'AMD leading NV' thing is nothing more than Koduri saying he believes they're ahead by several months, which might yet prove to be true.

 

[airfathaaaaa]

I don't know a single reputable source that has claimed that. Clickbait sites and forum warriors don't count.

In fact, the whole 'AMD leading NV' thing is nothing more than Koduri saying he believes they're ahead by several months, which might yet prove to be true.

they based it on the product samples on zauba.. we saw polaris 3-4 months before pascal therefore amd is 6 months ahead go figure
its like the ones that kept changing pascal launch from april to may and now they go for june

 

[tential]

So weren't we led to believe that Polaris was six months ahead of NV? it seems everyone got that wrong, NV knows to keep their secrets

Weren't we led to believe pascal has a 6 month lead ahead of Polaris? It seems everyone was wrong

 

[maddie]

I don't know a single reputable source that has claimed that. Clickbait sites and forum warriors don't count.

In fact, the whole 'AMD leading NV' thing is nothing more than Koduri saying he believes they're ahead by several months, which might yet prove to be true.

Too many ignore the complete sentence. Not surprised however.

Koduri:

We believe we’re several months ahead of this transition, especially for the notebook and the mainstream market

 

[Despoiler]

So weren't we led to believe that Polaris was six months ahead of NV? it seems everyone got that wrong, NV knows to keep their secrets

It was only Charlie @ S/A trolling with clickbait.

 

[antihelten]

Too many ignore the complete sentence. Not surprised however.

Koduri:

We believe we’re several months ahead of this transition, especially for the notebook and the mainstream market

In other words Koduri was basically talking about Polaris 10 vs. GP106, not the GP104 (which would be for the high end market, not the mainstream market).

 

[tential]

Too many ignore the complete sentence. Not surprised however.

Koduri:

We believe we’re several months ahead of this transition, especially for the notebook and the mainstream market

No, the whole context of that from question answering leads that statement to be confusing. It's definitely being taken out of context like no other when you read the full context it is vastly different

Also if Gp106 is coming later than gp104 and Polaris is 10 supposed to compete with Gp106 then it would be ahead of mainstream Nvidia offerings.

Just like Nvidia is ahead in high end offering...

 

[omek]

I think the "Order of 10" thing is neat, but if Nvidia is suggesting that Pascal is 10x faster than Maxwell, I'm skeptical to say the least. Curious to see what weird metric they use (like the "CEO Math" JH referenced last year) to arrive at that figure, if that's the case.

I'm sure it's to the order of 10 in wood screws, deep learning or something trivial/inconsequential.

This type of marketing is quite infantile. It's like giving a ferret something shiny, if it's shiny the ferret likes it even if it is a polished turd.


Threadcrapping and trolling are not allowed
Markfw900

 

[exar333]

We were told GDDR5X enters mass production in the summer. We're still in spring.

Either the timetable moved ahead significantly and no one told us, or NVidia and partners don't have chips to put on the boards yet, or at least don't have them in meaningful numbers.



The 200 series launched Dec 11 with retail availability Jan 12. Fury launched and no one could buy it due to HBM1 shortages for at least two months. That one feels familiar to the current memory situation.

Again, that was how many years ago? A month wasn't a HUGE deal there, and qualities were quite high once it was available. I wouldn't disagree though, but not terrible.

GDDRX may not be in 'mass production' but show a source that it isn't being produced? Only 1 SKU is proposed to have it, so yeah, it will be a niche offering.

 

[Jaydip]

Weren't we led to believe pascal has a 6 month lead ahead of Polaris? It seems everyone was wrong

I don't think Pascal ever had any set dates tbh, all we know came form leaks but it seems they were spot on!

 

[exar333]

Good for you. Doesn't change the fact the card was hard to come by for quite a while due to yield issues on HBM1.

Hard to come by <> paper launch.

Was the PS4 'paper launched'? No, just demand exceeded supply. Stop confusing the two.

 

[Jaydip]

It was only Charlie @ S/A trolling with clickbait.

I actually thought it was Koduri from RTG tbh but I have never read what he said in full except only bits or pieces here in AT.

 

[Jaydip]

I don't know a single reputable source that has claimed that. Clickbait sites and forum warriors don't count.

In fact, the whole 'AMD leading NV' thing is nothing more than Koduri saying he believes they're ahead by several months, which might yet prove to be true.

Unless it is a miracle I think for all intents and purpose NV is launching first but AMD will not be far behind for sure.

 

[AdamK47]

The beginning of June will be a very expensive time for me with Pascal and Broadwell-E releasing at the same time.

 

[railven]

The beginning of June will be a very expensive time for me with Pascal and Broadwell-E releasing at the same time.

I'm trying to talk myself out of buying Broadwell-E. My main CPU is going to hit two years and I feel like I have to upgrade it just out of stupidity.

Doubt the GF will OK a ~$1500 "upgrade" when she clearly sees me playing PSVita ports haha.

 

[Creig]

Well, I hope the "Order of 10" turns out better than their previous attempt of the "Power of 3", which flopped.

http://www.nvidia.com/object/powerof3.html

 

[96Firebird]

Well, I hope the "Order of 10" turns out better than their previous attempt of the "Power of 3", which flopped.

http://www.nvidia.com/object/powerof3.html

I'm sure it will.

 

[S.H.O.D.A.N.]

Unless it is a miracle I think for all intents and purpose NV is launching first

Based on what?