[Videocardz rumour] Vega pushed forward to October

[maddie]

Yep, its all based on a single random forum post without any information at all.

Seriously asking this.

How many times do you post something and refuse to provide any supporting facts?

Should we in the future, ignore everything you post like that?

 

[ShintaiDK]

Seriously asking this.

How many times do you post something and refuse to provide any supporting facts?

Should we in the future, ignore everything you post like that?

The information is already posted in the thread.

Not to mention you could just read the article linked in the first place. That goes from Videocardz to 3Dcenter.org to a random forum post. But that may be too much to ask, despite your deep interest in source criticism.

 

[maddie]

The information is already posted in the thread.

Not to mention you could just read the article linked in the first place. That goes from Videocardz to 3Dcenter.org to a random forum post. But that may be too much to ask, despite your deep interest in source criticism.

I'm not talking about this thread but you posting in general. You state something and when asked for supporting evidence, you often ignore the request.

Should we ignore your posts in those cases?

 

[zentan]

The information is already posted in the thread.

Not to mention you could just read the article linked in the first place. That goes from Videocardz to 3Dcenter.org to a random forum post. But that may be too much to ask, despite your deep interest in source criticism.

Hmm,for some reason double standards are followed by some.

I'm not talking about this thread but you posting in general. You state something and when asked for supporting evidence, you often ignore the request.

Should we ignore your posts in those cases?

Really? You ask a question he gave a direct answer now you resort to personal insults and trolling.
If you really need to ignore someone do you need his permission to use "ignore" button?

 

[maddie]

Hmm,for some reason double standards are followed by some.


Really? You ask a question he gave a direct answer now you resort to personal insults and trolling.
If you really need to ignore someone do you need his permission to use "ignore" button?

My intention is not to insult, and I don't see how I did, unless challenging someone is an insult. I try to read everyone as they all have something to contribute even if I disagree.

Posting something as a fact and refusing to say either, It's really my opinion or here is the supporting evidence, seems in poor taste.

 

[96Firebird]

Which is more likely, Vega moved up till October, or Polaris pushed back till October... :hmm:

 

[railven]

Which is more likely, Vega moved up till October, or Polaris pushed back till October... :hmm:

What if...Polaris is Vega???? Nah just messing.

So Polaris got delayed? If the Vega got pushed up rumor is being touted around here as a September release, welps, Polaris is going to come out after Vega?

The hell is going on at AMD HQ? GloFo? GloFo strikes again!

EDIT:

I think they just want to sell to the big spenders twice. If a 1080 can beat a 980ti, even by a little, some of those buyers will make the switch no matter the cost, then they'll line up again for the real high end in another 6 months.

Oh definitely. NV's got it's eye on the whales. If you actually think about it, the sell before new card method keeps you updated for a few dollars. I just sold my water cooled 980 Ti for $550. Thing cost me $625 after all the rebates and games. I lost $75 owning a 980 Ti for almost a year. That's a serious win to me. Pretty much paid for my 1080. (And if 1080 Ti/Vega 10 is impressive, I'll sell the 1080 and bump up, why not?)

 

[antihelten]

Which is more likely, Vega moved up till October, or Polaris pushed back till October... :hmm:

Or to rephrase it slightly: "Which is more likely, Vega moved up till October, or Polaris pushed back till October, considering that AMD has a large PR event in Macau later this month"

Holding a PR event just to announce the delay of their GPUs would be weird to put it mildly.

 

[railven]

Or to rephrase it slightly: "Which is more likely, Vega moved up till October, or Polaris pushed back till October, considering that AMD has a large PR event in Macau later this month"

Holding a PR event just to announce the delay of their GPUs would be weird to put it mildly.

Is the even related to Polaris/Vega/GPUs? Or are people just assuming that based on that one article?

Could be related to something else, perhaps that joint venture AMD has going with that Chinese company?

Soo many conflicting rumors right now. Unless AMD is trying to block NV and paper launch end of May with their cards allegedly delayed to October - that makes no sense.

 

[iiiankiii]

I reallly think Nvidia had the right approach to their gtx1080 launch. No tease, no speculation. Nvidia announced the Press Conference, introduced the cards, and stated a release date. The hype for the card is still fresh in people's head. No time for crazy speculations to ruin the actual performance. Nvidia kept quiet until they're ready to announce a release date. That's the way to go.

Unlike AMD, they kept on making the same mistakes. AMD kept on teasing the public about Polaris. They had a press conference, teased their performance BUT kept the release date in limbo. That's the wrong move. Given enough time, people will start to spread crazy rumors, start to expect crazy expection and then become sorely let down. Plus, the longer people have to wait, the quicker people will lose interest and move on to the next thing. We still don't have a definitive release date. That's horrible. AMD should haven't dragged the Press and the public through their PR stunts without giving a real release date. That's a no go.

 

[antihelten]

Is the even related to Polaris/Vega/GPUs? Or are people just assuming that based on that one article?

Could be related to something else, perhaps that joint venture AMD has going with that Chinese company?

Soo many conflicting rumors right now. Unless AMD is trying to block NV and paper launch end of May with their cards allegedly delayed to October - that makes no sense.

It is certainly possible that the Macau event is not centered around GPUs, but I very much doubt it has anything to do with the JV, since AMD normally wouldn't have big PR events for stuff like that.

Generally speaking AMD only does PR events like this for product launches or in relation to product launches.

If it is in fact Polaris, that might also explain the other rumor about OEM not being allowed to show of their Polaris products next door at Computex, since AMD would want to control the announcement of the new series, and make sure that it makes as big a splash as possible.

It is worth noting that AMD did something similar at last years Computex, where they didn't show off the new 300 series to the general public (only showing off some Fiji dies behind closed doors if I remember correctly). The official announcement of Fiji and the new 300 series didn't happen until the big AMD E3 event a few days later.

 

[guskline]

What if...Polaris is Vega???? Nah just messing.

So Polaris got delayed? If the Vega got pushed up rumor is being touted around here as a September release, welps, Polaris is going to come out after Vega?

The hell is going on at AMD HQ? GloFo? GloFo strikes again!

EDIT:


Oh definitely. NV's got it's eye on the whales. If you actually think about it, the sell before new card method keeps you updated for a few dollars. I just sold my water cooled 980 Ti for $550. Thing cost me $625 after all the rebates and games. I lost $75 owning a 980 Ti for almost a year. That's a serious win to me. Pretty much paid for my 1080. (And if 1080 Ti/Vega 10 is impressive, I'll sell the 1080 and bump up, why not?)

Railven: Did you sell the 980TI and the waterblock for $550 total or did you sell the waterblock separately? I'm not trying to be difficult but even with rebates $625 sounds right for the 980TI. A waterblock is another $100 minimum (I snagged mine for $120 including shipping). My experience with selling blocks is you are lucky to get 50% of the price. I'll add $60 to your total intake for $610.

Will you snag a GTX1080 for $599 or $699? You know the waterblock is going to be at least $120 ( I seriously doubt the 980TI block works). The cheapest you get away with is $719 ($599 +120) so if your total intake was $610 (gpu of $550 + 50% of block or $60) your outlay is at least $119. I suspect it will be closer to $219.

Water cooling isn't cheap, especially when the design changes!

 

[railven]

It is certainly possible that the Macau event is not centered around GPUs, but I very much doubt it has anything to do with the JV, since AMD normally wouldn't have big PR events for stuff like that.

Generally speaking AMD only does PR events like this for product launches or in relation to product launches.

If it is in fact Polaris, that might also explain the other rumor about OEM not being allowed to show of their Polaris products next door at Computex, since AMD would want to control the announcement of the new series, and make sure that it makes as big a splash as possible.

It is worth noting that AMD did something similar at last years Computex, where they didn't show off the new 300 series to the general public (only showing off some Fiji dies behind closed doors if I remember correctly). The official announcement of Fiji and the new 300 series didn't happen until the big AMD E3 event a few days later.

If I recall, AMD announced Fiji at Computex 2015 at their stage, not behind closed doors. As for cards, partners didn't have anything to show for Fiji since it was reference only. Fury wasn't expected for a few weeks of which only two partners got chips at the start.

Also, AMD was more vocal about their event, I remember reading about it in gaming sites not tech sites as it was the first PC-centric E3 presentation. This event seems only to be mentioned in one article.

Either way, my interest is peaked. Let's see what they got!

 

[railven]

Railven: Did you sell the 980TI and the waterblock for $550 total or did you sell the waterblock separately? I'm not trying to be difficult but even with rebates $625 sounds right for the 980TI. A waterblock is another $100 minimum (I snagged mine for $120 including shipping). My experience with selling blocks is you are lucky to get 50% of the price. I'll add $60 to your total intake for $610.

This was the EVGA Hybrid Kit, which they recently slashed the price on to $60. Some buyers tried to use that as a means to leverage me down, but I said I'm including the two shrouds that are available. Of it a buyer said he'd take it all so I was happy.

Will you snag a GTX1080 for $599 or $699? You know the waterblock is going to be at least $120 ( I seriously doubt the 980TI block works). The cheapest you get away with is $719 ($599 +120) so if your total intake was $610 (gpu of $550 + 50% of block or $60) your outlay is at least $119. I suspect it will be closer to $219.

Water cooling isn't cheap, especially when the design changes!

I'll be getting a non-founders reference PCB model (so possibly EVGA) and depending on the accoustics I'm not sure if I'll get a water block or use one of my current spare Corsair Hydros as a frankenstein (seeing how the EVGA kit attached, it's almost basically a standard CLC with 4 screws. Haha).

If it works, I'll probably do something similar to the GF's 1070.

Note: I don't have a custom loop, I work with CLCs cuz I'm too chicken haha.

 

[dark zero]

Better to sell all my shares of AMD since it will collapse very hard... Those GloFo... are trying to do magic tricks like summoning a genie on a bottle and are extremely mediocre... They must be sold to Samsung.

Also... All AMD cards are collapsing harder than Maxwell which is the best generation of nVIDIA ever seen.

Wondering if NVIDIA can sell the cards even at more expensive price since there are no competition and the old cards are history...

 

[antihelten]

If I recall, AMD announced Fiji at Computex 2015 at their stage, not behind closed doors. As for cards, partners didn't have anything to show for Fiji since it was reference only. Fury wasn't expected for a few weeks of which only two partners got chips at the start.

Also, AMD was more vocal about their event, I remember reading about it in gaming sites not tech sites as it was the first PC-centric E3 presentation. This event seems only to be mentioned in one article.

Either way, my interest is peaked. Let's see what they got!

It was a while ago so my memory is a bit flaky, so I just did a quick bit of googling. Basically AMD showed off a couple of non-functional Fiji dies at computex, but didn't announce the actual cards (i.e. the Fury X and its siblings) until the E3 event along with the rest of the 300 series lineup.

And you're probably right in your observation that this event seems much more low key than the E3 event (or the recent Capsaicin event for that matter).

 

[RussianSensation]

GloFo promises up to 2.41Ghz transistor switching speeds when moving from 28nm to 14nm. Whether or not the chip designer of the ASIC chooses higher IPC or higher clocks or lower power consumption or any combination (i.e., lower clocks + lower power usage + higher IPC) is up to them.

If Polaris 10 are only clocking at 850mhz, it's either a chip design failure (Fermi inter-connect fabric caused a 6 months delay), preliminary leaks with testing clocks [The baseline core clock of GK104's implementation will be similar to that of the GeForce GTX 480: 705 MHz, which clocks down to 300 MHz when the load is lowest, and the geometric domain (de facto "core") will clock up to 950 MHz on high load], base clock without boost activated, a tactic to confuse the competition/see who leaks data by purposely leaking to untrustworthy parties Polaris 10 at 850mhz, etc.

Initial samples of Kepler 670/680 were leaking 700mhz GPU clocks. Launch 680 easily hit 1110mhz on AT, or almost a 60% increase in GPU clocks from early leaked samples.

 

[MrTeal]

Heh, this rumour sure does get around. Fudzilla picked it up, and now it's stilling in my google news top stories feed.

 

[swilli89]

GloFo promises up to 2.41Ghz transistor switching speeds when moving from 28nm to 14nm. Whether or not the chip designer of the ASIC chooses higher IPC or higher clocks or lower power consumption or any combination (i.e., lower clocks + lower power usage + higher IPC) is up to them.

If Polaris 10 are only clocking at 850mhz, it's either a chip design failure (Fermi inter-connect fabric caused a 6 months delay), preliminary leaks with testing clocks [The baseline core clock of GK104's implementation will be similar to that of the GeForce GTX 480: 705 MHz, which clocks down to 300 MHz when the load is lowest, and the geometric domain (de facto "core") will clock up to 950 MHz on high load], base clock without boost activated, a tactic to confuse the competition/see who leaks data by purposely leaking to untrustworthy parties Polaris 10 at 850mhz, etc.

Initial samples of Kepler 670/680 were leaking 700mhz GPU clocks. Launch 680 easily hit 1110mhz on AT, or almost a 60% increase in GPU clocks from early leaked samples.

We are looking at either purposeful misguidance to throw nVidia off their scent or we are looking at a massive engineering problem.

14nm definitely allows for substantially higher clocked chips across the spectrum so a tiny Polaris 10@ 232mm that's not clocked at atleast 1.5ghz is a failure. We already saw a Polaris 10 playing Hitman at 1440p so I'm just very confused about what's going on..

 

[Headfoot]

I dont think I saw anyone post links to the actual source yet. It's just some dude in a forum. Unless he's supposed to be some known quantity, I've certainly got no reason to believe he knows more than any other forumite. http://www.forum-3dcenter.org/vbulletin/showthread.php?p=11027263#post11027263 and http://www.forum-3dcenter.org/vbulletin/showthread.php?p=11027319#post11027319

I've heard that Amd Vega brought forward to October 2016 ...

Release together / close with BF1. You'll take me for a fool act, but I can not tell you where the information comes but AMD seems to have been surprised himself by the clock speeds of Pascal. In SA there is a post that has to do with http://semiaccurate.com/forums/showpost.php?p=258604&postcount=69 first Rumor from my side. I stand for and we expect after the release order from (positive or negative)

I don't find this source very credible at all.

 

[BlitzWulf]

From a reddit poster an hour ago

https://www.reddit.com/r/Amd/commen...ris_validation_failed_might_launch_in/d32uwx7

I just finished listening/watching the quarterly stock market investors presentation from AMD. Basically, Lisa addresses her shareholder's with some information on how business is doing. During the presentation she stated that Polaris is still on track for it's original release date. I'd imagine this is the most current and up to date information, available only within the last 10 minutes.

Pretty vague on most topics really. To clarify a little, she actually said, "Polaris is still on track to be in the market this summer." I personally was hoping for spring, but still, kinda vague in her wording. She didn't delve too deeply into any one topic, as she was discussing semi custom, CPU's, and joint ventures as well.

seems it was an investors only stream

 

[xthetenth]

If Polaris 10 are only clocking at 850mhz, it's either a chip design failure (Fermi inter-connect fabric caused a 6 months delay), preliminary leaks with testing clocks [The baseline core clock of GK104's implementation will be similar to that of the GeForce GTX 480: 705 MHz, which clocks down to 300 MHz when the load is lowest, and the geometric domain (de facto "core") will clock up to 950 MHz on high load], base clock without boost activated, a tactic to confuse the competition/see who leaks data by purposely leaking to untrustworthy parties Polaris 10 at 850mhz, etc.

Initial samples of Kepler 670/680 were leaking 700mhz GPU clocks. Launch 680 easily hit 1110mhz on AT, or almost a 60% increase in GPU clocks from early leaked samples.

The thing that's throwing me is that if Polaris is only clocking at 850 MHz and they're expecting some huge increases from there, then either they've already gotten a few golden samples to do demos with, or the demos are showing phenomenal performance considering they're effectively significantly downclocked. It would make the SW demo make a lot of sense, using a card well below diminishing returns on clocks and undervolted like mad to turn in some fantastic power numbers, but the Hitman one seems like it doesn't fit unless they're pulling a similar stunt to the NV release trick of trying to conflate the VR niche case with the general case.

I reallly think Nvidia had the right approach to their gtx1080 launch. No tease, no speculation. Nvidia announced the Press Conference, introduced the cards, and stated a release date. The hype for the card is still fresh in people's head. No time for crazy speculations to ruin the actual performance. Nvidia kept quiet until they're ready to announce a release date. That's the way to go.

Unlike AMD, they kept on making the same mistakes. AMD kept on teasing the public about Polaris. They had a press conference, teased their performance BUT kept the release date in limbo. That's the wrong move. Given enough time, people will start to spread crazy rumors, start to expect crazy expection and then become sorely let down. Plus, the longer people have to wait, the quicker people will lose interest and move on to the next thing. We still don't have a definitive release date. That's horrible. AMD should haven't dragged the Press and the public through their PR stunts without giving a real release date. That's a no go.

I don't think AMD can do things the way NV did them. NV is the default. People weren't waiting for Polaris they were waiting for Pascal despite no news on either. AMD has to do things just to remind consumers that they exist. They do need a full reveal soon to get a burst of launch interest though.

 

[Headfoot]

The thing that's throwing me is that if Polaris is only clocking at 850 MHz and they're expecting some huge increases from there, then either they've already gotten a few golden samples to do demos with, or the demos are showing phenomenal performance considering they're effectively significantly downclocked.



I don't think AMD can do things the way NV did them. NV is the default. People weren't waiting for Polaris they were waiting for Pascal despite no news on either. AMD has to do things just to remind consumers that they exist. They do need a full reveal soon to get a burst of launch interest though.

I don't know why we'd believe that software written by people without access to a pre-release architecture would report the correct frequency. Realistically, we have no idea. 850 could be totally misreported, it could be a transitional state that gets reported, it could be any number of things because Polaris is unsupported by anyone but AMD internal tools and apparently AIDA64

 

[xthetenth]

I don't know why we'd believe that software written by people without access to a pre-release architecture would report the correct frequency. Realistically, we have no idea. 850 could be totally misreported, it could be a transitional state that gets reported, it could be any number of things because Polaris is unsupported by anyone but AMD internal tools and apparently AIDA64

Right. I'm discussing 850 more in the context of the delay rumor, where it would make a ton of sense in the context of GloFo sucking, but it doesn't seem to match with the demos.

 

[exar333]

The thing that's throwing me is that if Polaris is only clocking at 850 MHz and they're expecting some huge increases from there, then either they've already gotten a few golden samples to do demos with, or the demos are showing phenomenal performance considering they're effectively significantly downclocked. It would make the SW demo make a lot of sense, using a card well below diminishing returns on clocks and undervolted like mad to turn in some fantastic power numbers, but the Hitman one seems like it doesn't fit unless they're pulling a similar stunt to the NV release trick of trying to conflate the VR niche case with the general case.



I don't think AMD can do things the way NV did them. NV is the default. People weren't waiting for Polaris they were waiting for Pascal despite no news on either. AMD has to do things just to remind consumers that they exist. They do need a full reveal soon to get a burst of launch interest though.

This.

Polaris (as a 'mainstream' option) is great in theory but late. If they wanted to tap the VR demand, this should have been released in Q1 when everyone was building, ordering or prepping for VR. Instead, its launching after VR and after NV's 1080/1070. I do think a lot of people will just grab the 1070 as a 'minimum' VR card, or a used 390x/980/970 rather than wait for Polaris. On the other hand, if AMD can get this launched a few months before the 1060 is out, they could get some nice sales in the ~$200-250 range.

I almost expect NV to just hold onto their 1060/1060Ti and release it as a surprise at the same time to spoil the Polaris launch. Would NOT be surprised at all.