From ATI to AMD back to ATI? A Journey in Futility @ [H]

AMD is tied to GF, they have to get a certain amount of wafers from them otherwise it breaks their contract and the fines are quite hefty. So if GF is having issues with their process, AMD is stuck and have to weather the storm.

yes but given that i spoke to him specifically about polaris, his reply seemed carefully worded to be ambiguous, problems with the process, or problems with the design ? or both ?
 
Polaris not so bright [Hardocp]From ATI to AMD back to ATI? A Journey in Futility


I must say, you have managed to stir things up a lot Kyle :)


I wish for more things to be stirred up, lol. Kyle, let the butthurt flow through you! Write more articles!!! ROFL.

Industry needs to be rocked for a bit.
 
yes but given that i spoke to him specifically about polaris, his reply seemed carefully worded to be ambiguous, problems with the process, or problems with the design ? or both ?


I would put it more on the process side of things. But we will know soon enough.
 
At the end of the day its AMDs call as to who goes. Just like it would be for Nvidia. I mean at the end of the day who pays for airfare and room and board? Because whoever is paying should be the one that gets to decide.

Sure it is AMD's decision and they're well within their rights but as consumers, it also signals to us that they're scared of their product or event being critiqued and want to control the flow of information. By not sending out an invitation to a well established site like this one, it seems they only want yes men at their events. You know there's a word for that, it's called insecure.
 
Sure it is AMD's decision and they're well within their rights but as consumers, it also signals to us that they're scared of their product or event being critiqued and want to control the flow of information. By not sending out an invitation to a well established site like this one, it seems they only want yes men at their events. You know there's a word for that, it's called insecure.
Their is a word for what you are doing as well. Its called Guessing.
I do not really think anyone expects Polaris to compete with a 1080. It would be a nice surprise but its just not going to happen.
 
At the end of the day its AMDs call as to who goes. Just like it would be for Nvidia. I mean at the end of the day who pays for airfare and room and board? Because whoever is paying should be the one that gets to decide.

Kyle and Brent have reviewed several 380xs, 280x, 380s, 960s etc

The same way they have reviewed 295X2 etc etc, the fact is AMD's PR department is acting pissy because Kyle upset Roy Taylor. Kyle is free to be as big a mean, lean bag of dicks (sorry Kyle) as he wants to be; he's an independent reviewer. AMD is one of the world's biggest electronics companies. It is ridiculous that they cut [H] off with the pretense that these are not for HardOCP's audience.
 
One thing is for sure. Kyle's butt is on the line if Polaris is actually decent.

I hope Polaris is a piece of shit so I don't feel bad about the pair of R9 290s I recently acquired.

I didn't see him say anywhere in that article that Polaris is a piece of shit. Rather it clearly states, "In the simplest terms AMD has created a product that runs hotter and slower than its competition's new architecture by a potentially significant margin." So to me that isn't really surprising because Polaris 11 is a low power GPU and Polaris 10 is only targetting midrange. Now if Polaris 10 is $300 and barely keeps up with a 390x, then AMD might be in a bit of trouble considering the 1070 is at $380.

Their is a word for what you are doing as well. Its called Guessing.
I do not really think anyone expects Polaris to compete with a 1080. It would be a nice surprise but its just not going to happen.

As stated above, I was never under the delusion that Polaris would be a GP104 competitor. I expect Vega and it's derivatives will do that job in October (hopefully).
 
Kyle and Brent have reviewed several 380xs, 280x, 380s, 960s etc

The same way they have reviewed 295X2 etc etc, the fact is AMD's PR department is acting pissy because Kyle upset Roy Taylor. Kyle is free to be as big a mean, lean bag of dicks (sorry Kyle) as he wants to be; he's an independent reviewer. AMD is one of the world's biggest electronics companies. It is ridiculous that they cut [H] off with the pretense that these are not for HardOCP's audience.
I never claimed they didnt review AMD cards did I? Hell my other post even says they do so fairly.
At the end of the day AMD still can do as they please when it comes to paying out the ass for people to come and review hardware. I do think they are making a stupid mistake but I really do not think anyone can disagree with the premise that its AMD choice on who goes. Not the other way around.
 
Those Anand fuckers were probably the community voting for that other girl in that baseball poll Kyle posted.

Sad article either way. Does Intel even have a reason to be involved in the dedicated GPU category? Seemed like their interest with Larabee fell off fast.
 
It's in AMD's best interest to always send an invite to [H] for the benefit of the readers/community of [H] - regardless of whos name it is AMD has to write out on the invitation. Instead - by blacklisting [H] they are sending the message that the AMD owners in the [H] forums/AMD owners who read [H] reviews - or potential AMD owners of the same group(s) - don't matter to AMD - whether that was AMD's intention or not.

That's why it's just bad practice to any company to take things so personal and be so petty like this.
 
Last edited:
Those Anand fuckers were probably the community voting for that other girl in that baseball poll Kyle posted.

Sad article either way. Does Intel even have a reason to be involved in the dedicated GPU category? Seemed like their interest with Larabee fell off fast.

Larabee was a joke and it was a different time then. The Intel of today very much cares about PC gaming and is focused on it.
 
I didn't see him say anywhere in that article that Polaris is a piece of shit. Rather it clearly states, "In the simplest terms AMD has created a product that runs hotter and slower than its competition's new architecture by a potentially significant margin." So to me that isn't really surprising because Polaris 11 is a low power GPU and Polaris 10 is only targetting midrange. Now if Polaris 10 is $300 and barely keeps up with a 390x, then AMD might be in a bit of trouble considering the 1070 is at $380.



As stated above, I was never under the delusion that Polaris would be a GP104 competitor. I expect Vega and it's derivatives will do that job in October (hopefully).

I didn't claim that Kyle said Polaris is a piece of shit.

I want it to be one so I don't feel about my recent purchase.

I have the right to be selfish :D
 
Well if cards are given out for reviewers to plow through and test and give feedback to AMD before any NDA, that would be better and kinda prudent for a process for a good hard launch. If NDA is late June, AMD would be keeping their promise of mid year launch - well hopefully with cards that can be bought. GloFlo 30% yield issue with Apple maybe carried over to AMD, maybe AMD should just go to Samsung and give GloFlo the middle finger if they are the ones once again FU.

If HardOCP is late on reviews, then we will just go elsewhere and probably just bring it back here in the end for discussion. I also think that once HardOCP actually can get some good samples that you will probably see the best in debt review on the net as well. Sometimes it is not bad if last if you have the best in the end.
 
yes but given that i spoke to him specifically about polaris, his reply seemed carefully worded to be ambiguous, problems with the process, or problems with the design ? or both ?
Leldra, I have a big chinese porcelain Pink elephant in my office. Would you believe that ?
 
I still hold out hope for Zen. Latest rumors were the clocks were better than expected, and it might actually go toe to toe with Skylake, instead of being relegated to compete with Maxwell like it previously was rumored.

FWIW, Skylake is an Intel CPU, Maxwell is an nVidia GPU. Possibly you meant Haswell-E, as that's an Intel CPU. I wish Zen well, but the 8c16t 5960X TOTL HW-E is like setting up the heavyweight champ for your first fight. It's strong at stock, and 50% OCS are not difficult to achieve. If Zen can get Skylake IPC and OC ability in an 8c HT package, it'll be more than a little surprising.
 
[H] has painted themselves into a corner.

When Polaris drops, if it performs well and is a significant improvement in Perf/W, then [H] and it's "sources" "data" and declared "non-bias" are laid out for all to see.

[H] has become one step closer to WTF, Tech!?!? with this article (based on scuttle-butt/rumors/water-cooler whispers, etc) and could only have out-WTF-Tech!?-ed them if this article had been sourced from a poster in some other non [H] tech forum.

Now the online tech community gets to see for ourselves how this plays out with respect to Polaris and it's performance, and where [H] really stands on the bias/non-bias and "privy knowledge" versus "pulled some load from a diaper and flung it at the screen" spectrum.

[H] might have some info correct about internal politics, potential Intel deals, or the like and may be entirely incorrect on Polaris.

Looking forward to how this all plays out.
 
[H] might have some info correct about internal politics, potential Intel deals, or the like and may be entirely incorrect on Polaris.
As of two weeks ago, I am dead on about Polaris. And I do not think AMD is going to fix the problem by seeding reviewers with cherry picked samples of the cards this weekend. I do not think they can get back on track that quickly with their clocks for production. Polaris was never meant to be a middle of the stack part, but you can spin it however your want.
 
[H] has painted themselves into a corner.

When Polaris drops, if it performs well and is a significant improvement in Perf/W, then [H] and it's "sources" "data" and declared "non-bias" are laid out for all to see.

[H] has become one step closer to WTF, Tech!?!? with this article (based on scuttle-butt/rumors/water-cooler whispers, etc) and could only have out-WTF-Tech!?-ed them if this article had been sourced from a poster in some other non [H] tech forum.

Now the online tech community gets to see for ourselves how this plays out with respect to Polaris and it's performance, and where [H] really stands on the bias/non-bias and "privy knowledge" versus "pulled some load from a diaper and flung it at the screen" spectrum.

[H] might have some info correct about internal politics, potential Intel deals, or the like and may be entirely incorrect on Polaris.

Looking forward to how this all plays out.


All that on what it means 'if they're not right' and only a sentence that says 'they might also be right'. Lol, YOUR bias is showing.
 
Intel acquiring a real GPU company that then gets help from intels engineers and direct access to intels top of the line fabs would be fucking amazing. This would cause nvidia's asshole to seriously pucker hard.

This is wishful thinking. About the only thing the acquisition of RTG offers to Intel is a way to make their IGPs slightly better. AIB cards are too small of a market for Intel.
 
Those Anand fuckers were probably the community voting for that other girl in that baseball poll Kyle posted.

Sad article either way. Does Intel even have a reason to be involved in the dedicated GPU category? Seemed like their interest with Larabee fell off fast.

I have many friends on Anandtech and a few places, that was a bit extreme, I try not to be overly judgemental myself.

Posters vary there, like any site.
 
Yeah, they always fail to realize Nvidia will still have to compete with itself, and will still need to be bringing money in from new sales. You do that by making new products people want to buy at prices people want to buy them at. This mythical 'entry level GPU for $500' beast would not do that, and is nothing but fanboi fear mongering.

Perhaps you forgot that because the 970 didnt have a real competitor, nvidia didnt bothered in lowering the price due to demand?

And on the same boat, since it was the best card to get, we sucked it up and bought them, myself included and price be dammed.

So yes, they would do it if they can and looks like they will be able to.
 
Pretty much agree. Kyle has placed enough faith into "talk" to put his site and personal reputation on the line with this article. If it does not pan out the way he predicts it will be a serious blow to [H]. This is why journalists do not go to press on just "talk."

Let us be frank about this. Unless Kyle has got a hold of someone's mouth that stands next to Lisa or Koduri during negotiations then it is all RUMOR with god knows how many degrees of separation to his ear.

At this point it is just a discussion about what may be. When the time comes and passes will we see if this article was prophetic or a true disservice to the community in which it serves.

[H] has painted themselves into a corner.

When Polaris drops, if it performs well and is a significant improvement in Perf/W, then [H] and it's "sources" "data" and declared "non-bias" are laid out for all to see.

[H] has become one step closer to WTF, Tech!?!? with this article (based on scuttle-butt/rumors/water-cooler whispers, etc) and could only have out-WTF-Tech!?-ed them if this article had been sourced from a poster in some other non [H] tech forum.

Now the online tech community gets to see for ourselves how this plays out with respect to Polaris and it's performance, and where [H] really stands on the bias/non-bias and "privy knowledge" versus "pulled some load from a diaper and flung it at the screen" spectrum.

[H] might have some info correct about internal politics, potential Intel deals, or the like and may be entirely incorrect on Polaris.

Looking forward to how this all plays out.
 
Leldra, I have a big chinese porcelain Pink elephant in my office. Would you believe that ?
I don't believe you have an office.
jon-stewart-oh-snap.gif
 
The funniest part about the people stating Kyle is not a good tech journalist is that these same people are not journalists or writers themselves. Most of them can barely type a complete thought.

If someone like Anand Lal Shimpi calls out Kyle on his journalistic integrity, I might start to take notice.
 
I have many friends on Anandtech and a few places, that was a bit extreme, I try not to be overly judgemental myself.

Posters vary there, like any site.

I have a friend named Anand, and he is a douche bag.

Don't have any negative feeling towards Anandtech however.
 
Doubt we'll see any confirmation of that for many years to come. But whether it was meant to be a mainstream card or not, if it's priced right, it could do extremely well. If AMD can afford to sell it at that price, that is.

If they price it at >$300, for a GPU with 390x-class performance, they're sunk.
And I think I have stated this more than once, Polaris positioning is all about price after they figure out what clocks production silicon is going to support.
 
There are other professional review sites being included. AdoredTV just managed to get fans and makes videos that have information. I dont particularly care for his videos but he's not the only one invited. [H] is just considered extremely biased.



its funny how u guys dont even realize u are as much fanboys as you think others are. Any sensible person would take this article with a grain of salt. Given history and lack of sources etc.

I guess you didn't read the first comment in the thread. Hint: it was from me.

Oh, one more thing. I have nothing against AMD at all unlike lots of the AMD uber alles crew have against NV. I'll buy either company if they give me the performance I'm looking for. What I do have a problem with is emotional children that think AMD (or NV for that matter) somehow are morally superior and should be emotionally supported for some odd reason. At the moment AMD fanbois are the target of my ire since they are the most vocal and foolish at this time.
 
Last edited:
Perhaps you forgot that because the 970 didnt have a real competitor, nvidia didnt bothered in lowering the price due to demand?

And on the same boat, since it was the best card to get, we sucked it up and bought them, myself included and price be dammed.

So yes, they would do it if they can and looks like they will be able to.

So, if price be damned (as you say, verbatim) why didn't you buy the Titan X or a 980 (or 980Ti if it was out by then/when you bought yours)? Or was it because the 970 was a competitive card at a competitive price for what the product was, from Nvidia - which you are saying Nvidia will not do? Nice contradiction.
 
I think by now its pretty clear what it was made for which by design makes it an affordable mid range part.
 
If AMD is truly cherry picking samples for reviewers, then Launch date might not be 29th of June, only release of info and initial reviews, lets see what happens.

Also they might not have hit their intended performance doesn't mean they are far away though even 5% of a miss can put them from high mid range/low performance end to low midrange, they might be able to get away with that by increases power usage.

And when was the last time [H] reported on rumors like so many other rumor sites? If [H] reports on rumor, they say its rumor lol very clearly.
 
Kyle, please tell us you are writing a made-for-TV movie script based on all of this! :LOL:

I'm one of the old schoolers who has been around for eons and remember all the way back to when ATI / NVIDIA / 3dfx started their wars. This current stuff is all really crazy shit. :eek:
 
PrMinisterGR;5280327 said:
The funniest part are the bunch of illiterates on his forum, congratulating him for being a "good reporter".

Meanwhile, in the RTG the Great Leader is preparing to sink the company by getting more contracts:
TOUxBv3.jpg

Let me put it this way, put this angry sounding article aside for a minute.

2560 SPs, 1250MHz (I'm being conservative) = 6.4Tflops

Compare with Fury 3584 @ 1000MHz = 7.17Tflops
That's only 10% more.

980Ti 2816 @ 1200 = 6.75tflop

Notice how 980Ti scores higher than Fury ?

Polaris 10 should have 160TMUs and 64ROPs if I'm not mistaken


It's geometry performance is claimed to be doubled, accounting for the higher clocks + primitive discard we should be looking at Maxwell levels of polygon throughput

By all means it should be doing better than the Fury overall. Should be doing as well as a 980Ti really.

Maybe it's good enough for a macbook, but the 3dm scores leaked are very underwhelming, same with AotS scores

Naturally this is nothing to be happy about
 
you guys fail to realize you will get less and less. Even if there are options at different prices, they can milk for the most profit with the least performance gains. a 970 card in a generation would cost $600. Sure you could get a card that is decent for $300, but you would not get as good as you would have with competition. When you are a monopoly you create the landscape.

This idea that they would have to compete with themselves is also silly. If you are the only option, you think that is as big a factor?
This is not the same market as the 90's. A company that flaunts their monopoly usually ends up in one of two spots, in court (the EU is really good at this one) or with a new stronger competitor who exploited the market they ignored. For good or bad the GPU market is price segmented. There are lots of buyers for cards in the $100-200 range and fewer as the price goes up. Intel can't fully exploit integrated graphics while there are cheap cards competing with them. Without those, Intel would crank up the performance of their IGPs just a tad to satisfy the low end and OEM market (a big part of the market) and let NVidia scramble to hold onto a shrinking mid and high end market, if they price themselves out of people's budget. Ultimately a graphics card (especially a high end one) is a luxury item that not all gamers need. So, yes, NVidia would still need to compete with itself or its monopoly would cease to exist very very quickly.
 
Back
Top