Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

GTX 1080 is here

SEANMCADSEANMCAD Member EpicPosts: 16,775
http://www.theverge.com/circuitbreaker/2016/5/7/11615806/nvidia-gtx-1080-1070-pascal-specs-price-release-date

I didnt see this coming...

this soon

but yeah, now that many if not all the new titles require a GTX 970 it should be to anyone surprise that this was also right around the corner. Pretty much same pattern as always on hardware deployments 

Please do not respond to me, even if I ask you a question, its rhetorical.

Please do not respond to me

«13456

Comments

  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    The May 27 date is basically an announcement of when they'll start taking pre-orders.  It's not at all similar to a date of when you can actually get a card and play games on it.  If you really think that cards with GDDR5X are going to have real commercial availability less than two months after the first GDDR5X chips started sampling, I've got some ocean-front property in Kansas to sell you.
  • jesteralwaysjesteralways Member RarePosts: 2,560
    I guess they don't wanna take the risk with what AMD is about to release soon. Like vultures they took results of AMD's HBM tech research and then improved their own product. I have never been more ashamed to say that i used to be a Nvidia user.

    Boobs are LIFE, Boobs are LOVE, Boobs are JUSTICE, Boobs are mankind's HOPES and DREAMS. People who complain about boobs have lost their humanity.

  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
  • Doug_BDoug_B Member UncommonPosts: 153
    edited May 2016
    Looks like a great card for the past. Still using ddr ram is kind of a let down since pascal was slated for HMB . I like AMD's gpu designs better and will wait for the second gen HBM cards.
    Bachelor's in Web Design and Multimedia
  • jesteralwaysjesteralways Member RarePosts: 2,560
    edited May 2016
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..

    Boobs are LIFE, Boobs are LOVE, Boobs are JUSTICE, Boobs are mankind's HOPES and DREAMS. People who complain about boobs have lost their humanity.

  • Doug_BDoug_B Member UncommonPosts: 153
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    Bachelor's in Web Design and Multimedia
  • MalaboogaMalabooga Member UncommonPosts: 2,977
    edited May 2016
    Hmmm, NVidia avoiding 980ti in all their PR materials lol (comapring 1080 either to 1000$ Titan X or 400$ 980). Its marginally slower than Titan X and has been at 530-550$ for a while now. 1080 is 10-30% faster and 699$ at start, 599$ later.

    1080 is kinda not very convincing so far.

    1070 (439$ ->379$) much more interesting, but no perf data and will probably compete with Polaris 10 in similar price range so it reamains to be seen when both are released with benchmarks and retail prices.
  • SEANMCADSEANMCAD Member EpicPosts: 16,775
    edited May 2016
    posts make me laugh...

    I had said not but about 1 month ago that its likely the GTX970 will be a standard card with many games stating it as suggested requirements  AND that the next generation of video cards for the consumer market would be in full swing before the end of this year AND that when that starts to happen you will see a lot of good deals on GTX970

    I guess I was wrong

    Please do not respond to me, even if I ask you a question, its rhetorical.

    Please do not respond to me

  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    HBM2 is a JEDEC standard.  Anyone who wants to use it is free to do so.  This isn't just AMD and Nvidia.  For example, Altera has announced that they're going to use HBM2 in FPGAs:

    http://newsroom.altera.com/press-releases/nr-dram-sip.htm

    AMD did it that way intentionally, as if it were proprietary, it wouldn't have nearly as large of volume and would be much more expensive.  HBM was co-developed by AMD and Hynix, and I can assure you that Hynix would have been a lot less interested if they thought AMD would be the only customer they could ever get.  How much does it cost to buy XDR2, again?
  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
  • maple2maple2 Member UncommonPosts: 161
    THese new cards will be nice. but ye we will not see HBM 2 til 1080 ti
  • Doug_BDoug_B Member UncommonPosts: 153
    edited May 2016
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    Bachelor's in Web Design and Multimedia
  • maple2maple2 Member UncommonPosts: 161
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    u really miss alot of brain dont u? they never stated that Pascels Non flagship will use HBM 2.. only the Titan class and Ti version. while 1080 will use GDDR5X. which is faster then GDDR5. and 1070 will use GDDR5. the 256 bit is enough for 4k gaming with 8g vram. nvidia is smart enough to use the tech to their outmost limit.

    amd only release all their cards at once because they dont know how to marketing.. and their HBM was a fat failur. it brought nothing performence was horrible and the 200 series overheated so badly. and uses way to much power. on top of that amd cards clock speed is year 2000. while nvidia just pushed to a 1600+ core speed and insanely fast memory's. Nvidia has and always been the best. just look at the new tech they brought in.
  • ceratop001ceratop001 Member RarePosts: 1,594
    edited May 2016

    TV and the Internet are good because they keep stupid people from spending too much time out in public.

    – Douglas Coupland


     
  • Doug_BDoug_B Member UncommonPosts: 153
    edited May 2016
    maple2 said:
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    u really miss alot of brain dont u? they never stated that Pascels Non flagship will use HBM 2.. only the Titan class and Ti version. while 1080 will use GDDR5X. which is faster then GDDR5. and 1070 will use GDDR5. the 256 bit is enough for 4k gaming with 8g vram. nvidia is smart enough to use the tech to their outmost limit.

    amd only release all their cards at once because they dont know how to marketing.. and their HBM was a fat failur. it brought nothing performence was horrible and the 200 series overheated so badly. and uses way to much power. on top of that amd cards clock speed is year 2000. while nvidia just pushed to a 1600+ core speed and insanely fast memory's. Nvidia has and always been the best. just look at the new tech they brought in.
    Thanks for your opinion.  http://wccftech.com/nvidia-pascal-trouble/
    People are going to be seeing a big change in market shares between AMD and nvidia.
    Bachelor's in Web Design and Multimedia
  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    AMD decided that their high end parts with HBM2 would nominally be a separate architecture from their mid-range and low-end parts with GDDR5.  The former is Polaris and the latter is Vega.  Nvidia decided that their high end parts with HBM2 and their mid-range and low-end parts with GDDR5/GDDR5X would all be called Pascal.  That's just a difference in how they name things, not in what they're doing.
  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    maple2 said:
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    u really miss alot of brain dont u? they never stated that Pascels Non flagship will use HBM 2.. only the Titan class and Ti version. while 1080 will use GDDR5X. which is faster then GDDR5. and 1070 will use GDDR5. the 256 bit is enough for 4k gaming with 8g vram. nvidia is smart enough to use the tech to their outmost limit.

    amd only release all their cards at once because they dont know how to marketing.. and their HBM was a fat failur. it brought nothing performence was horrible and the 200 series overheated so badly. and uses way to much power. on top of that amd cards clock speed is year 2000. while nvidia just pushed to a 1600+ core speed and insanely fast memory's. Nvidia has and always been the best. just look at the new tech they brought in.
    Well if you want to compare Nvidia GPUs from 2017 to AMD GPUs from 2012, then of course the former are better.  That's a difference between 2017 and 2012, not between Nvidia and AMD.

    Considering that it's probable that Polaris will be available before Pascal, you could at least try comparing Pascal to Polaris or GCN to Maxwell.  Actually, if you want to talk about the 200 series, you should compare GCN to Kepler, not Maxwell, with only the exception of Tonga, which came later.

    You should also be aware that AMD's released cards with HBM are in the Fury series, not the 200 series, and don't have a number at all in the model name.
  • maple2maple2 Member UncommonPosts: 161
    edited May 2016
    Quizzical said:
    maple2 said:
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    u really miss alot of brain dont u? they never stated that Pascels Non flagship will use HBM 2.. only the Titan class and Ti version. while 1080 will use GDDR5X. which is faster then GDDR5. and 1070 will use GDDR5. the 256 bit is enough for 4k gaming with 8g vram. nvidia is smart enough to use the tech to their outmost limit.

    amd only release all their cards at once because they dont know how to marketing.. and their HBM was a fat failur. it brought nothing performence was horrible and the 200 series overheated so badly. and uses way to much power. on top of that amd cards clock speed is year 2000. while nvidia just pushed to a 1600+ core speed and insanely fast memory's. Nvidia has and always been the best. just look at the new tech they brought in.
    Well if you want to compare Nvidia GPUs from 2017 to AMD GPUs from 2012, then of course the former are better.  That's a difference between 2017 and 2012, not between Nvidia and AMD.

    Considering that it's probable that Polaris will be available before Pascal, you could at least try comparing Pascal to Polaris or GCN to Maxwell.  Actually, if you want to talk about the 200 series, you should compare GCN to Kepler, not Maxwell, with only the exception of Tonga, which came later.

    You should also be aware that AMD's released cards with HBM are in the Fury series, not the 200 series, and don't have a number at all in the model name.
    eh? i never compared any Gpu's... i said Amd's 200 series and even 300 series is a flat out failur with far to high powerdraw. and far to unstable drivers. etc.

    and also polaris has no release date while pascel is out in just 20 days... with 1080 and 1070 later. and 1080 ti most likely end of 2016. while amd will come out when? in 2018.
  • maple2maple2 Member UncommonPosts: 161
    Doug_B said:
    maple2 said:
    Doug_B said:
    Quizzical said:
    Doug_B said:
    Quizzical said:
    There's nothing wrong with Nvidia adopting HBM.  The idea of industry standards is that it's cheaper if everyone uses the same few types of memory than if every memory vendor or GPU vendor has to create their own memory standards from scratch.  That allows you to, for example, buy the same DDR3 memory modules regardless of whether you're buying a CPU from Intel or AMD, at least if you get a CPU from the DDR3 era.  That's a good thing.
    For nearly 10 years AMD researched HBM, they could not even push out a decent graphics card that is compatible with everything since they were determined to perfect HBM, while NVIDIA used the already established and working tech to churn out one after another uber compatible product and took huge share of market. And now that AMD has almost perfected HBM and brought out HBM 2, nvidia just take it and use it to make their products perform better, again!! If you don't see the wrong in this, then it is just sad..
    Pascal is not a HBM gpu. Just saying.
    The cards announced yesterday don't use HBM, but the top end GP100 chip does use HBM2.
    That comes out when ? ... AMD released their entire line in one showing.. the 290 and 290x with working HBM.. Nvidia, nope , just one card. if the rumor is true, you won't see Nvidia using HBM until 2017 -18 simply because they can't get it to work..that is kind of what happens when nvidia doesn't innovate, they are left in the dust trying to copy and paste.

    u really miss alot of brain dont u? they never stated that Pascels Non flagship will use HBM 2.. only the Titan class and Ti version. while 1080 will use GDDR5X. which is faster then GDDR5. and 1070 will use GDDR5. the 256 bit is enough for 4k gaming with 8g vram. nvidia is smart enough to use the tech to their outmost limit.

    amd only release all their cards at once because they dont know how to marketing.. and their HBM was a fat failur. it brought nothing performence was horrible and the 200 series overheated so badly. and uses way to much power. on top of that amd cards clock speed is year 2000. while nvidia just pushed to a 1600+ core speed and insanely fast memory's. Nvidia has and always been the best. just look at the new tech they brought in.
    Thanks for your opinion.  http://wccftech.com/nvidia-pascal-trouble/
    People are going to be seeing a big change in market shares between AMD and nvidia.
    trouble? another useless information from the so called wccftech.. Pascel has never been in trouble.. and never will lol..
  • Doug_BDoug_B Member UncommonPosts: 153
    edited May 2016
    http://arstechnica.com/gaming/2016/01/amd-confirms-high-end-polaris-gpu-in-development-for-2016/

    More innovation : http://arstechnica.co.uk/gaming/2016/01/amds-new-graphics-architecture-is-called-polaris/

    THe pascal is a dinasour and everything they showed the nvidia video was driver/software based. Same old design from nvidia trying to pass it off as something "new". nvidia is sweating bullets, they are two steps behind AMD right now.
    Bachelor's in Web Design and Multimedia
  • QuizzicalQuizzical Member LegendaryPosts: 25,347
    maple2 said:
    Quizzical said:
    Well if you want to compare Nvidia GPUs from 2017 to AMD GPUs from 2012, then of course the former are better.  That's a difference between 2017 and 2012, not between Nvidia and AMD.

    Considering that it's probable that Polaris will be available before Pascal, you could at least try comparing Pascal to Polaris or GCN to Maxwell.  Actually, if you want to talk about the 200 series, you should compare GCN to Kepler, not Maxwell, with only the exception of Tonga, which came later.

    You should also be aware that AMD's released cards with HBM are in the Fury series, not the 200 series, and don't have a number at all in the model name.
    eh? i never compared any Gpu's... i said Amd's 200 series and even 300 series is a flat out failur with far to high powerdraw. and far to unstable drivers. etc.

    and also polaris has no release date while pascel is out in just 20 days... with 1080 and 1070 later. and 1080 ti most likely end of 2016. while amd will come out when? in 2018.
    You should try reading the thread before replying sometime.  The GTX 1080 with GDDR5X is not going to be widely available in May.  Or June.  Or July, for that matter.  Polaris, on the other hand, probably will be available for you to buy one by the end of July.  Traditionally, GPUs don't announce their launch date ahead of time, but keep everything under NDA, and then suddenly they're available to buy the day that the vendor announces the specs.

    Since you brought up 2018, by then, AMD will probably have Navi ready, presumably on a 10 nm process node.  It's going to be better than Pascal.  But not necessarily better than Volta, which will be the real competition for it.

    If you want to accuse most of AMD's GCN GPUs of having too high of a power draw, then you might want to note that the contemporary Nvidia Kepler GPUs had about the same power draw.  AMD's drivers are about as good as Nvidia's and have been for many years.
  • holdenhamletholdenhamlet Member EpicPosts: 3,772
    I guess they don't wanna take the risk with what AMD is about to release soon. Like vultures they took results of AMD's HBM tech research and then improved their own product. I have never been more ashamed to say that i used to be a Nvidia user.
    LOL.  Uh no.  It's the beauty of competition.  Thanks to competition we get great products from both vendors.
  • MalaboogaMalabooga Member UncommonPosts: 2,977
    edited May 2016
    I guess they don't wanna take the risk with what AMD is about to release soon. Like vultures they took results of AMD's HBM tech research and then improved their own product. I have never been more ashamed to say that i used to be a Nvidia user.
    LOL.  Uh no.  It's the beauty of competition.  Thanks to competition we get great products from both vendors.
    No, thanks to AMD and Hynix who didnt keep HBM tech for themselves you get great products from both vendors (even some FPGA vendors will use HBM)

    They could have gone the route like NVidia and dont share it with anyone and NVidia would still be stuck on GDDR5.
  • Doug_BDoug_B Member UncommonPosts: 153
    I guess they don't wanna take the risk with what AMD is about to release soon. Like vultures they took results of AMD's HBM tech research and then improved their own product. I have never been more ashamed to say that i used to be a Nvidia user.
    LOL.  Uh no.  It's the beauty of competition.  Thanks to competition we get great products from both vendors.
    Really isn't a competition when Nvidia can't even get last years tech to work....
    Bachelor's in Web Design and Multimedia
  • holdenhamletholdenhamlet Member EpicPosts: 3,772
    edited May 2016
    Malabooga said:
    I guess they don't wanna take the risk with what AMD is about to release soon. Like vultures they took results of AMD's HBM tech research and then improved their own product. I have never been more ashamed to say that i used to be a Nvidia user.
    LOL.  Uh no.  It's the beauty of competition.  Thanks to competition we get great products from both vendors.
    No, thanks to AMD and Hynix who didnt keep HBM tech for themselves you get great products from both vendors (even some FPGA vendors will use HBM)

    They could have gone the route like NVidia and dont share it with anyone and NVidia would still be stuck on GDDR5.
    Wait so Nvidia was supposed to not use a good new technology because they didn't invent it?

    How does that benefit anyone?

    Video card vendor fanboys...

    Besides, you're acting like AMD never copied anything from Nvidia, which is not true.
Sign In or Register to comment.