NVIDIA Volta with GDDR6 in early 2018?

[FoxEye]

Just buy a card now. With the tech market no matter when you buy something better will come along within 6 months.

If you have a set budget within which you always keep, then something new comes along approx every 18 months. Eg mid-ranger gen n-1 to mid-range gen n.

The only time what you said is true is if you always want the very best card, in which case nV will happily release a new Titan (etc) every 6 months

 

[AthlonXP1800]

SAMSUNG starts producing industry’s first 16-Gigabit GDDR6 memory
https://videocardz.com/press-releas...ucing-industrys-first-16-gigabit-gddr6-memory

Look like Samsung is the first company to start mass production industry's first 16 gigabit GDDR6 memory likely will use in upcoming GTX 1080 Ti successor Volta or Ampere 2080 Ti due to launch at GTC 2018 in March.​

 

[KillBoY_UK]

2080 ti announced in march? Unlikey, a "2080" a big fat maybe.

Ether way any new cards best to buy FAST due to mining

 

[Samwell]

I bet it'll start with a new Titan at GTC, which will be more gaming focused than Titan V and faster than TV. Then April for a 2080, 2080Ti will take more time. Before April Nvidia can't launch anything in quantities because of chinese new year. Companies tend to start production after this, as too many people are in vacation in chinese new year.

 

[Cooper]

2080 ti announced in march? Unlikey, a "2080" a big fat maybe.

Yeah, they're not going to announce any replacement for the 1080Ti at the Volta/Ampere unveiling, that will be for the 1070/80 replacements only, 2080Ti will be a 2019 release.

 

[ToTheMax]

Yeah, they're not going to announce any replacement for the 1080Ti at the Volta/Ampere unveiling, that will be for the 1070/80 replacements only, 2080Ti will be a 2019 release.

2080 sometime in Q2 I recon, inventory needs to build up for GDDR6 graphic card production.

 

[Kaapstad]

Don't think we will see any HBM gaming cards from NVidia now that GDDR6 is available.

I hope AMD also start using GDDR6 for their future cards.

 

[Cooper]

2080 sometime in Q2 I recon, inventory needs to build up for GDDR6 graphic card production.

March: 2070/2080 announcement, May/June availability

September/October: 2060 announcement & immediate availability (using duff 2070/2080 parts)

Jan/Feb 2019: 2080Ti announcement, March/April availability

 

[Killer7xx]

March: 2070/2080 announcement, May/June availability

September/October: 2060 announcement & immediate availability (using duff 2070/2080 parts)

Jan/Feb 2019: 2080Ti announcement, March/April availability

If miners have their way, you won't be seeing any real availability. Volta is a killer when it comes to GPU mining.
https://hothardware.com/news/nvidia-titan-v-volta-ethereum

Unless Nvidia gimps them, these will be snapped up in seconds after someone reveals their hashrates with the most common cryptocurrencies.

 

[Cooper]

If miners have their way, you won't be seeing any real availability. Volta is a killer when it comes to GPU mining.
https://hothardware.com/news/nvidia-titan-v-volta-ethereum

I prefer not to think about those scumbags TBH, it just raises my blood pressure thinking about the long term damage those wastes of space are doing to the PC industry.

 

[ToTheMax]

I prefer not to think about those scumbags TBH, it just raises my blood pressure thinking about the long term damage those wastes of space are doing to the PC industry.

Agree, but why can't the manufactures just double/quadruple capacity to meet the demand? It's not as if this trend came out of nowhere, it's been known for years.

 

[Freddie1980]

Agree, but why can't the manufactures just double/quadruple capacity to meet the demand? It's not as if this trend came out of nowhere, it's been known for years.

I believe AMD did increase it's order when Bitcoin was a thing for GPU miners several years ago, the trouble was when the market collapsed the 2nd market was flooded with cheap 290/290X's and AIB and retailers were left with large inventories which resulted in AMD slashing prices to clear stock and have something to compete with the GTX970.

The problem is to ramp up production is something you can't do overnight, some reports I've read say it can take up to 3 months for making the decision for the the increased units to find there way into the hands of retailers by which time the mining craze may have crashed again, it's a very risky decision.

 

[d_brennen]

Often production of the GPU die ends WAY before the product goes EOL too. Should they run out unexpectedly (or should demand remain constant longer than a conservative estimate) it can be a costly and lengthy process to get a production run set up with the fab

 

[nashathedog]

Don't think we will see any HBM gaming cards from NVidia now that GDDR6 is available.

I hope AMD also start using GDDR6 for their future cards.

Please God, Make AMD drop HBM, Amen.

I think I'm gonna have to start adding this in to my getting ready for bed routine.

 

[KillBoY_UK]

Please God, Make AMD drop HBM, Amen.

I think I'm gonna have to start adding this in to my getting ready for bed routine.

I can't see this happening.

AMD's cores badly need the power savings (to budget more to the core) and the extra memory bandwidth vega's design is staved of it), AMD GPU need R300 or a Tesla like shift in design to fix this and get back on top.

Which I hope Navi will be so they can use more cost effective GDD5X/GDDR6

 

[nashathedog]

I can't see this happening.

AMD's cores badly need the power savings (to budget more to the core) and the extra memory bandwidth vega's design is staved of it), AMD GPU need R300 or a Tesla like shift in design to fix this and get back on top.

Which I hope Navi will be so they can use more cost effective GDD5X/GDDR6

I'm hoping Navi has 5x/6 across the range, I'm pretty sure using HBM was the cause of the Vega release delays and why there's a shortage even now so they need to focus on what made Polaris what it is and what went wrong to lead to Vega being what that is. Polaris may be slower than Vega but it's a lot better than Vega when you consider the power draw & heat. A bigger Polaris chip would have been a better option than Vega and if they want to have a hope of catching back up with Nvidia they need to make changes, not stick to their guns and continue flogging the dead horse..

 

[Rroff]

Often production of the GPU die ends WAY before the product goes EOL too. Should they run out unexpectedly (or should demand remain constant longer than a conservative estimate) it can be a costly and lengthy process to get a production run set up with the fab

IIRC - its like a 45 day turn around just for the production run. (EDIT: Can take as much as ~90 days for a complex design).

 

[Kaapstad]

I can't see this happening.

AMD's cores badly need the power savings (to budget more to the core) and the extra memory bandwidth vega's design is staved of it), AMD GPU need R300 or a Tesla like shift in design to fix this and get back on top.

Which I hope Navi will be so they can use more cost effective GDD5X/GDDR6

There is no reason to stop a GDDR5(X)(6) card from using a bit more power than a HBM based one. There are also advantages too in moving the memory away from the GPU core as it makes cooling easier and higher clockspeeds more possible.

If the next round of AMD cards do use GDDR6 and consume an extra 20 watts who cares if it means the cards are easier to manufacture and there are plenty of them on the shelves.

 

[KillBoY_UK]

There is no reason to stop a GDDR5(X)(6) card from using a bit more power than a HBM based one. There are also advantages too in moving the memory away from the GPU core as it makes cooling easier and higher clockspeeds more possible.

If the next round of AMD cards do use GDDR6 and consume an extra 20 watts who cares if it means the cards are easier to manufacture and there are plenty of them on the shelves.

It all depends, 20w on a top tier 250w cards is not much in the grand scheme of things but when your memory controller and ram is sucking up 1/3 of the boards TDP like with RX480/RX580 it makes you wonder how much performance is being held back to get inside the typical board TDP's of 250/150/75w.

I thing it be better for AMD to come out with a 5970 type product, ie top quite taking the performance crown but 90-95% of the cutting edge performance on a smaller cheaper to produce chip which much less power consumption.

Its happened in that past when nvidia (NV30/Fermi) or AMD (Rage128/R600/Vega) have come out with big hot power hungry stinkers they had to start from scratch to get back in the "game" so to speak.

 

[Davedree]

I'm hoping Navi has 5x/6 across the range, I'm pretty sure using HBM was the cause of the Vega release delays and why there's a shortage even now so they need to focus on what made Polaris what it is and what went wrong to lead to Vega being what that is. Polaris may be slower than Vega but it's a lot better than Vega when you consider the power draw & heat. A bigger Polaris chip would have been a better option than Vega and if they want to have a hope of catching back up with Nvidia they need to make changes, not stick to their guns and continue flogging the dead horse..

Scaling up the 3072 shader Polaris was axed because of AMd's antiquated gcn core and limited 4 wide geometry engines. It could only scale so far before hitting memory bandwidth and power consumption problems vs performance.
The 384 bit memory controller and bandwidth would have brought marginal gains over the Polaris 10 (2304 shader). Whilst 6gb was too little and 12gb would further upset power consumption and price, where a Gp104 8gb would have totally smashed it.
Where did they salvage the dies, well they are in the underwhelming Xbox Scorpio at 40/44cu.
And had to use hbm2, but unfortunately the reality was it was just a redrawn Fury II that arrived late.

Gp104 caught AMD with their pants down.