Mysterious Graphics Chip from Nvidia, the ‘JM601’ GPU, Spotted in Shipping Manifest on Zauba

[Gregster]

Needless to say we have never encountered this nomenclature before. The naming JM601 is completely alien to me. In the past, GPUs popping up in the shipping manifests usually had their actual names. Considering that the previous Nvidia chip spotted also had a weird name (the 433×3, a possible scrambled name for the upcoming Dual GPU), there are two possible things that could be happening: 1) Nvidia has finally wizened up and started scrambling the code names of the GPUs to hide their progress from our eyes or 2) the nomenclature is correct and this is something we have never seen before.

It is worth noting here that chips spotted at Zauba usually become public in a matter of few months (usually 4-5). Since this coincides with the release date of Pascal – it is a possibility that we are looking at a flagship chip of the 16nm FinFET generation.

Read more: http://wccftech.com/nvidia-graphics-chip-jm-601-gpu-spotted/#ixzz3pgKCGe9h

Interesting and could be the flagship pascal chip coming our way soon. I do hope so anyways and can't wait to try out the new node

 

[tommybhoy]

My guess is no1. but with baby Pascal much like baby Maxwell which was a runaway success, but who knows.

 

[Rroff]

Mobile "Pascal" part under 9 series branding.

 

[P.B]

a new high end mobile part would be nice hopefully faster than the mobile 980 and that looks a beast as it is.

 

[Cooper]

Probably someones lunch box.

 

[humbug]

Probably someones lunch box.

 

[nashathedog]

Low end Pascal with gddr5 or HBM1, If your lucky.

 

[Rofflay]

Well of course they are going to try it on with another 20% and GDDR5x in my opinion before we get anywhere near this proper Pascal. They said the same thing about Maxwell too OMG HUGE JUMP.

The above would only wet me if it had DP 1.3.

 

[Klunt Bumskrint]

The first triple gpu card. The Titan X3

 

[sidthesexist]

The first triple gpu card. The Titan X3

And to power it...

 

[CAT-THE-FIFTH]

Mebbe a pipecleaner Pascal part??

 

[Gregster]

28nm has been with us far too long and I think even if you are Red and Green biased, we can all agree on that. Bring on the 16nm scrap

 

[Boomstick777]

28nm has been with us far too long and I think even if you are Red and Green biased, we can all agree on that. Bring on the 16nm scrap

Far far far far to long..

April would be good for high end die shrunk cards. Going to skip the first round of wallet busters, not really interested in the GDDR5X cards if they come first. I'l hold out for the proper high end Ti / Titan with HBM 2.0 whenever they arrive..

 

[Gregster]

Yer same and HBM 2 please.

 

[CAT-THE-FIFTH]

Even a 40% to 50% improvement would put a £200 to £300 14NM/16NM cards at around GTX980TI performance levels.

 

[LambChop]

I'd rather have the latest die shrink Ti with gddr5x. hbm is talking a good game but delivered nothing on the fury. Wither hbm 2 is any better, who knows.

 

[Boomstick777]

I'd rather have the latest die shrink Ti with gddr5x. hbm is talking a good game but delivered nothing on the fury. Wither hbm 2 is any better, who knows.

It delivered massively, much lower power consumption, enabled small form factors to have a big punch. Huge memory bandwidth and more efficient than GDDR5, i.e less amount of HBM can do the same task than more GDDR5.

HBM 2.0 looks even better from rumored spec, twice the speed and option for higher capacity. So on the Nvidia side where they are already power efficient with HBM coming to the table the performance per watt will be awesome, along with die shrink and new architecture this is all good stuff.

Ti will have HBM 2.0 as that is the future for the high end. The lower spec cards would be where GDDR5X could be utilized as it's cheaper probably in much more supply and can allow more supply of HBM to be reserved for the higher end cards, while HBM is still in constrained supply and expensive.

 

[Mei]

these big gfx cards will look a bit like dinosaurs when hbm becomes the norm
we might see some new case designs too, stubby atx cases

 

[bru]

Well the last 55nm card was the 285, two nodes latter (40nm -28nm) the first full fat card was the Titan, which was about THREE times faster than the 285. So the first 16nm card (two nodes down from the currant 28nm cards) could in theory be 3 times faster than what we have now.

Now of course I do not expect the first Pascal cards to be 3 times faster than a TitanX, but I really do think we will be seeing a much bigger leap than the suggested 50%.

 

[Dalandius]

Looks like interesting times are ahead!!