G.Skill TridentZ5 RGB DDR5 7200 CL34 2x16 GB review
ASUS TUF Gaming B760-PLUS WIFI D4 review
Netac NV7000 2 TB NVMe SSD Review
ASUS GeForce RTX 4080 Noctua OC Edition review
MSI Clutch GM51 Wireless mouse review
ASUS ROG STRIX B760-F Gaming WIFI review
Asus ROG Harpe Ace Aim Lab Edition mouse review
SteelSeries Arctis Nova Pro Headset review
Ryzen 7800X3D preview - 7950X3D One CCD Disabled
MSI VIGOR GK71 SONIC Blue keyboard review
Rumor: GeForce GTX 1080 8GB gddr5x based on Pascal GPU in May
Well, at least I have a nice juicy rumor just in time for the weekend. Asian based Benchlife claims that Nvidia will release a Pascal architecture based GeForce GTX 1080 with 8GB gddr5x in May. It would make sense as HBM2 will be available in volume only later in the year.
There's a precise date being mentioned as well, May 27th, four days prior to the Computex exhibition. Obviously Nvidia is not going to comment on this rumor.
The GP104 based GTX 1080 would get a 256-bit memory controller tied to 8GB of gddr5x memory. The usage of gddr5x over regular gddr5 is rather speculative as it isn't even in volume production that we know of? Micron (the only one that will fab these gddr5x ICs) announced in a recent blog post that they intend to commence high-volume production of GDDR5X chips in mid-2016, or sometime in the summer. But who knows, Nvidia have gotten their hands on an early batch for a couple of thousand cards right ? GDDR5X is your standard GDDR5 memory however, opposed to delivering 32 byte/access to the memory cells, this is doubled up towards 64 byte/access. And that in theory could double up graphics card memory bandwidth.
Early indications according to a presentation show numbers with the memory capable of doing up-to 10 to 12 Gbps, and in the future 16 Gbps. So your high-end graphics cards these days hover at say 400 GB/s. With GDDR5X that could increase to 800~1000 GB/sec and thus these are very significant improvements, actually they are competitive enough with HBM.
Pascal GPUs are fabbed in the Taiwan Semiconductor Manufacturing Company (TSMC) (recently hit by an earthquake) based on the new 16nm FinFET process. 16nm should result in significant power savings. The card would only need one 8-pin connector for power needs. The card would get a DVI, HDMI and two display ports. From the looks of things, and as expected .. this is not 'Big Pascal' GP100. That GPU would get 17 billion transistors made on the 16 nm FinFET+ node at TSMC and likely the HBM2 memory.
So to recap, in the latest prognosis Nvidia are rumored to launch the first high-end GP104 GPUs with models GTX 1080 and GTX 1070 (should Nvidia use this naming scheme). These chips would come in a 37.5×37.5mm BGA package, and get 2152 pins.
LimitbreakOr
Senior Member
Posts: 500
Joined: 2015-04-02
Senior Member
Posts: 500
Joined: 2015-04-02
#5245084 Posted on: 03/14/2016 08:59 PM
I don't think it would be tied into the engine radiator. Lots of higher end cars have supplementary radiators for all kinds stuff. In fact I'm pretty sure some cars even now have watercooled ECU's.
http://www.dana.com/~/media/danacom/files/media-asset/ptg--to-upload/brochure/hybrid-and-electric-vehicle-cooling-products.pdf
I guess I did say tie in, I just meant that into the overall design.
Okay, that's alot more reasonable lol
I don't think it would be tied into the engine radiator. Lots of higher end cars have supplementary radiators for all kinds stuff. In fact I'm pretty sure some cars even now have watercooled ECU's.
http://www.dana.com/~/media/danacom/files/media-asset/ptg--to-upload/brochure/hybrid-and-electric-vehicle-cooling-products.pdf
I guess I did say tie in, I just meant that into the overall design.
Okay, that's alot more reasonable lol
PrMinisterGR
Senior Member
Posts: 8099
Joined: 2014-09-27
Senior Member
Posts: 8099
Joined: 2014-09-27
#5245089 Posted on: 03/14/2016 09:12 PM
You dont need to refresh all the menory for every frame, thats like saying that every time you hammer in a nail, you need to clear the entire room of equipment and reload all the equipment for the next nail.
To give you a better example of how it works, look at the xBone; it has a lowly ddr3 memory with 68Gb/s, so to compensate, they used a huge part of the apu for 32mb cache that has 204 GB/s(102 In/102 Out). Essentially, that tiny 32mb enables the xBone to run nicely at 900p.
Thanks for that, I'll look on how the XboxOne (and the 360 that had something similar) are doing it.
They didn't present pascal at an auto show. They presented PX2 at CES. And they've been talking about cars for years, hence why it's called a PX2 and not a PX1, this isn't a "new focus". They said during the conference that all the demos were running on a Titan X. They also said they wouldn't have working PX2 units into the hands of partners until Q2 2016 and wide-scale release until 2H 2016. He mentioned Pascal like twice during the entire conference, saying that the GPU's would be used on the unit and the relative performance they expected out of the entire unit and that's about it -- it was all stuff they covered when they first announced Pascal back in 2015.
You realize there are cars in places like Libya, where the hottest recorded temperature was 134F? Or the fact that cars already have liquid cooling systems that can easily just tie into the PX2? Or the fact that this is targetted at being 300w which is probably a bit more than your standard ECU? And where does it say that Pascal is 150W? or it's relative performance in comparison to 980Ti? They gave the TDP/Tflops of all of PX2, which includes the two Parker CPU's and the A57s. Drawing conclusions about Pascal from the PX2 is a bit of a stretch.
I actually did a bit of research about the consumption of ECUs. I found this IEEE paper which is about power over ethernet on ECUs, but it has an indication on the power consumption of the average ECU. Whatever NVIDIA is proposing seems to have ~3-5x times higher energy requirements than the largest current systems.
That could explain the need for more robust cooling. I can't understand how you are defending them about the PX2 chipset issue though. They said that it was the PX2 board but it was a 980m board instead. They could have simply said that this was the form factor they were targeting. But that would have meant admitting that Pascal wasn't there as even a placeholder yet, which would put them at least a quarter behind AMD in any actual hardware shipment.
You dont need to refresh all the menory for every frame, thats like saying that every time you hammer in a nail, you need to clear the entire room of equipment and reload all the equipment for the next nail.
To give you a better example of how it works, look at the xBone; it has a lowly ddr3 memory with 68Gb/s, so to compensate, they used a huge part of the apu for 32mb cache that has 204 GB/s(102 In/102 Out). Essentially, that tiny 32mb enables the xBone to run nicely at 900p.
Thanks for that, I'll look on how the XboxOne (and the 360 that had something similar) are doing it.
They didn't present pascal at an auto show. They presented PX2 at CES. And they've been talking about cars for years, hence why it's called a PX2 and not a PX1, this isn't a "new focus". They said during the conference that all the demos were running on a Titan X. They also said they wouldn't have working PX2 units into the hands of partners until Q2 2016 and wide-scale release until 2H 2016. He mentioned Pascal like twice during the entire conference, saying that the GPU's would be used on the unit and the relative performance they expected out of the entire unit and that's about it -- it was all stuff they covered when they first announced Pascal back in 2015.
You realize there are cars in places like Libya, where the hottest recorded temperature was 134F? Or the fact that cars already have liquid cooling systems that can easily just tie into the PX2? Or the fact that this is targetted at being 300w which is probably a bit more than your standard ECU? And where does it say that Pascal is 150W? or it's relative performance in comparison to 980Ti? They gave the TDP/Tflops of all of PX2, which includes the two Parker CPU's and the A57s. Drawing conclusions about Pascal from the PX2 is a bit of a stretch.
I actually did a bit of research about the consumption of ECUs. I found this IEEE paper which is about power over ethernet on ECUs, but it has an indication on the power consumption of the average ECU. Whatever NVIDIA is proposing seems to have ~3-5x times higher energy requirements than the largest current systems.
That could explain the need for more robust cooling. I can't understand how you are defending them about the PX2 chipset issue though. They said that it was the PX2 board but it was a 980m board instead. They could have simply said that this was the form factor they were targeting. But that would have meant admitting that Pascal wasn't there as even a placeholder yet, which would put them at least a quarter behind AMD in any actual hardware shipment.
Reardan
Senior Member
Posts: 596
Joined: 2014-09-21
Senior Member
Posts: 596
Joined: 2014-09-21
#5245096 Posted on: 03/14/2016 09:18 PM
Yeah I don't think you'd want it in the same loop with the engine block, but certainly car manufacturers are accustomed to intercoolers and would be easily capable of engineering the device in. Some cars even have secondary loops for other things and some electric cars cool the battery. For those applications, it's possible the PX2 could just be integrated in. For regular cars, it would need its own minor accommodations.
Just want to point out, tying a electronic component to the engines radiator wouldn't do it any favors, car rads reach boiling temperature like 230f*.
Yeah I don't think you'd want it in the same loop with the engine block, but certainly car manufacturers are accustomed to intercoolers and would be easily capable of engineering the device in. Some cars even have secondary loops for other things and some electric cars cool the battery. For those applications, it's possible the PX2 could just be integrated in. For regular cars, it would need its own minor accommodations.
Denial
Senior Member
Posts: 14045
Joined: 2004-05-16
Senior Member
Posts: 14045
Joined: 2004-05-16
#5245106 Posted on: 03/14/2016 09:33 PM
I'm not defending them about that. They should have said that it was a mock-up. I just don't think it's fair to conclude that PASCAL isn't ready based on it. I don't know if they have working silicon or not. I don't know if that mock-up unit was designed a year ago and they just took it because it was readily available. I imagine that if they told people they are shipping units in Q2, that they at least planned on having it ready by Q2. But who knows, maybe this is another Fermi and they trying to save face before the impending launch disaster.
That could explain the need for more robust cooling. I can't understand how you are defending them about the PX2 chipset issue though. They said that it was the PX2 board but it was a 980m board instead. They could have simply said that this was the form factor they were targeting. But that would have meant admitting that Pascal wasn't there as even a placeholder yet, which would put them at least a quarter behind AMD in any actual hardware shipment.
I'm not defending them about that. They should have said that it was a mock-up. I just don't think it's fair to conclude that PASCAL isn't ready based on it. I don't know if they have working silicon or not. I don't know if that mock-up unit was designed a year ago and they just took it because it was readily available. I imagine that if they told people they are shipping units in Q2, that they at least planned on having it ready by Q2. But who knows, maybe this is another Fermi and they trying to save face before the impending launch disaster.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 14045
Joined: 2004-05-16
I don't think it would be tied into the engine radiator. Lots of higher end cars have supplementary radiators for all kinds stuff. In fact I'm pretty sure some cars even now have watercooled ECU's.
http://www.dana.com/~/media/danacom/files/media-asset/ptg--to-upload/brochure/hybrid-and-electric-vehicle-cooling-products.pdf
I guess I did say tie in, I just meant that into the overall design.