AMD Radeon RX 480 Supplies at Launch

[Ieldra]

The most common argument I see is the 'amd offers unmatched performance per dollar', does it?

$240 on the 8GB 480 vs $380 on the 1070.

If the 480 offers less than 60% the performance of the 1070 it will actually offer less performance per dollar, leaving AotS data aside, the firestrike leaks show the 1070% being 50% faster if I'm not mistaken

 

[Ieldra]

You have to remember that AMD claimed this card was never meant to compete with the 1080 (not sure about the 1070, but seems doubtful). So two different targets.

That said, I'd expect quite a few people will be looking to purchase a pair of 480s at launch so they can outperform a 1080 like the AMD slides showed.

Unfortunately for AMD, the 1080 performs around 10% better in reality, you can check the benchmark browsers on AotS. I matched the fps they claimed the 1080 gets, on a 980ti

 

[FSRedWing]

Tell me what would you do if you could make 200 million in the performance segment in one quarter vs 100 million in the mainstream segment? Which would you target, All of this when your company is in crushing debt and negative cash flow? Keeping in mind once you make a choice you loose the rest of the market not chosen for the next year, there are no take backs.

Its a fantasy question YOU ARE THE CEO OF AMD for this question, what would you do?

That depends on a number of factors:

1) How does your plan for achieving long-term profitability look like? Even when you're bleeding money, it can still be the better option to go for market share in the short term.
2) Can you produce enough products to satisfy demand? If not, those 200 million might well turn into less than 100 million.
3) How does your product pipeline look like? In AMD's case, they have Vega following Polaris in relatively short order. Does it make sense to cannibalise the sales of a future product in the performance segment when you could cover the mainstream segment instead with your current one?

 

[lolfail9001]

Unfortunately for AMD, the 1080 performs around 10% better in reality, you can check the benchmark browsers on AotS. I matched the fps they claimed the 1080 gets, on a 980ti

Did you downclock CPU tho to stock 5930k (well, and dropped down memory)?

 

[razor1]

That depends on a number of factors:

1) How does your plan for achieving long-term profitability look like? Even when you're bleeding money, it can still be the better option to go for market share in the short term.
2) Can you produce enough products to satisfy demand? If not, those 200 million might well turn into less than 100 million.
3) How does your product pipeline look like? In AMD's case, they have Vega following Polaris in relatively short order. Does it make sense to cannibalise the sales of a future product in the performance segment when you could cover the mainstream segment instead with your current one?

We are talking about the prospect of the rx480 as the performance part if Pascal wasn't around or Pascal's performance was a bit lower that it is. I should have added this.

1) so by forgoing the performance segment and taking in less cash so the competition can beat you again, and again for another year in the very segment that has highest volume sales and most margins is a good thing when you look at long term profitability when you can't gain marketshare?

2) You don't think AMD can produce enough stock of rx480?

3) Vega is inconsequential to this argument because if P10 was low end performance, Vega is high end performance and enthusiast. (pretty much going after the ti versions, and Titan)

 

[harmattan]

Actually find the 390x [H] article where Kyle responds to the discussion that the next line of cards is very special. Even AMD thought they had a winner back then but it seems they had to adapt hence what you saw at GDC 2016.

Actually, I recall similar quotes earlier last year from quite a few other sites hinting that AMD had something special coming in their next line-up and we might want to hold off on 390/x and Fury. Seems a lot of people thought (or were fed) earlier last year that there was a performance card coming this year.

While I'm interested academically and in a macro sense the history of this chip - and I suspect [H] will be able to write a page-turner retrospective on all this a couple months - I'm more interested in seeing how 480 performs at this point.

 

[Deleted member 93354]

Some other things we don't know, who is having an easier time with production? tsmc with those 314mm dies? or GF with the samsung assist with 14nm? Even if nvidia clearly takes the perf per watt and perf per mm categories, they will radically lose out on the most important factor of all for some time. perf/dollar.

majority of gamers are still on 1080p displays, do they get a much more widely available 480 that runs their games at 60+fps for around 200-230, or a 1070 that runs them at 80fps for twice the price, where they can't even see the increased frames on their 60Hz display. This is not even a contest. Take your performance crown, much like the protoss zealot, strong and proud, more expensive to make and produce, but stronger as a singular unit as well. polaris is the zerg swarm. And based off good enough performance to allowing gamers to thrive at 200 bucks and a seemingly greater ability to produce cards off samsungs 14nm process and lower die size strategy, nvidia will be overwhelmed. Kyle and razor and leidra will consider this a failure, the nvidia cards are technically superior! So was beta max.

That 1070, so strong! so proud! so dead

I think we can ALL agree that the RX480 is going to be the champ around $200->$250 without question for typical gaming.

There is some very large concern though about AMD's ability to scale up to 1070/1080 levels for things like complex VR, and 1440p or 4K gaming.

It's a heck of a lot easier to SCALE down to AMD's RX480 level once you have the hard designs done than it is to scale up. So NVIDIA could jump in quick with a entire top to bottom line. The question is whether they can undercut AMD's aggressive pricing.

 

[razor1]

They don't need to undercut AMD to have a successful product in the midrange, they can just have the same price at performance bump. All the while possibly even keeping the same TDP There are many way to do battle.

 

[Rvenger]

Nvidia can charge $50 more and have lesser performance and it will outsell since the TDP will likely be lower thats how it will be marketed. Look at the 750ti.

Key is marketing. AMD has none.

 

[NKD]

Looks like amd has enough stock and lot to go around. Seriously why the fuck one would want something otherwise. If amd can create competition that's good for all of us. This crap about oh it's a failed product cuz it was suppose to be high and bla bla, it's anyone's guess. But there is no issue with performance for he price they are asking. There is so much damn hype around it reading in forums and what not. People are ready to click that buy buttom. For 229 for 8gb model and OC it to 1400 and get fury performance. That's easy money to spend right there. Who cares about how much profit amd is making. Cuz I don't. If they are going for market share looks like they will sure get some cuz people are ready to buy this shit. Price is too good for this performance bracket. As far as the card running hot and in efficient that shit is already debunked. It's using a cheap ass cooler with copper core and aluminum fins and still staying in low 70s. Yea too hot I guess, smh.

If nvidia can give me similar performance for y dollar they have my money but they don't. So I could care less who's name is on the box if it's saving me money. Those wanna spend 450, 699 you got 1070 and 1080. Be happy, and game on.

 

[ChosenUno]

NVIDIA can also outmarketing AMD with regards to VR since AMD doesn't seem to have made any progress software-wise, while NVIDIA with their various new improvements to VR rendering will likely whoop same-tier AMD cards in VR performance, provided those rendering techniques make it into VR games.

I really hope AMD has techs similar to SMP. Fingers crossed.

 

[NKD]

NVIDIA can also outmarketing AMD with regards to VR since AMD doesn't seem to have made any progress software-wise, while NVIDIA with their various new improvements to VR rendering will likely whoop same-tier AMD cards in VR performance, provided those rendering techniques make it into VR games.

I really hope AMD has techs similar to SMP. Fingers crossed.

No one is going to implement VR if it can't get to the masses. Trust me it will start with consoles first and then you will have those ports come over, guess who owns the console space? Yea amd. Reason for Polaris. Ps4 neo, next Xbox and now cheap card for PCs. Anyone who thinks this wasn't their plan a year ago is mislead.

 

[Algrim]

I posted this in another thread. I think the reason why people here and other forums I've visited are not so enthused is because they already are in the performance sector (i.e. 290x, 390x, Fury X, 980 Ti, Titan X) and can't really find an upgrade to what they have. People with 980 Tis and some Titans are overclocking to where the 1080 is overclocked as well. Why would this segment be excited about a 199/239 (or the 'beast' AIB cards in the 260+ region) card that gives them less performance than what they already have?

For someone like me, currently limping along with my iGPU until I find the card I like, this new card is great! It reduces the price point of admission to a performance segment I thought was not achievable. Whether or not nVidia has a competitor to it is not really an issue. When the RX 480 is released on June 29th, if it measures up to expectations, it's going to invalidate any current card under a Fury X AMD-side or a 1070 nVidia side. At least, in my own opinion.

 

[Rvenger]

All a matter of perspective. I love the performance of my 295x2 and would love to have that same performance without the 500w TDP. So I am going to attempt 2x 480s

 

[harmattan]

All a matter of perspective. I love the performance of my 295x2 and would love to have that same performance without the 500w TDP. So I am going to attempt 2x 480s

Coming from someone with a 295x2 and enough experience with Crossfire over the years to know what to expect, you can expect performance with dual 480s with an ugly, giant asterisk.

 

[ppi]

Well, when you read this comment, and couple that with the fact that Raja in the interview with VentureBeat mentioned Polaris 10 being a "premium graphics" part, is it that hard to make the logical conclusion? Of course AMD had to be vague, they'd be stupid to outright say whatever, because at that point they had no idea what Pascal was capable of.

Unless he was bulshitting with respect to what is "premium graphics", AMD had to be extremely bullish on the clock speed. If we take 480@1266~390X, they would need 1800ish clocks to match 980Ti, and exceed by some margin 2GHz in order to meaningfully beat it.

Could they have had such an expectation?

 

[Chimpee]

Unless he was bulshitting with respect to what is "premium graphics", AMD had to be extremely bullish on the clock speed. If we take 480@1266~390X, they would need 1800ish clocks to match 980Ti, and exceed by some margin 2GHz in order to meaningfully beat it.

Could they have had such an expectation?

My personal opinion, I think AMD original want to target Fury X performance and phase out the Fury line completely since it cost too much for their liking.

 

[tybert7]

I posted this in another thread. I think the reason why people here and other forums I've visited are not so enthused is because they already are in the performance sector (i.e. 290x, 390x, Fury X, 980 Ti, Titan X) and can't really find an upgrade to what they have. People with 980 Tis and some Titans are overclocking to where the 1080 is overclocked as well. Why would this segment be excited about a 199/239 (or the 'beast' AIB cards in the 260+ region) card that gives them less performance than what they already have?

For someone like me, currently limping along with my iGPU until I find the card I like, this new card is great! It reduces the price point of admission to a performance segment I thought was not achievable. Whether or not nVidia has a competitor to it is not really an issue. When the RX 480 is released on June 29th, if it measures up to expectations, it's going to invalidate any current card under a Fury X AMD-side or a 1070 nVidia side. At least, in my own opinion.

It really is a case of tesla model s class drivers looking at the pending model 3 with turned up noses. What does THAT do for ME ?!??? And as we all know, the only people that exist in the world, is themselves. Actually, no, I think most tesla model s drivers see the model 3 as a great thing, which is why the shade and bile thrown here at amd being "forced" to lower themselves into the 200 dollar gutter is so discordant with other analysis. It is inconceivable that this price point is a feature and not a bug. It's a feature if amd wants to claw back some marketshare.

rx 460 ~100 dollars
rx 470 ~ 150
rx 480 ~200-229 for 8GB - 300 for variations of scaled up performance with overclocks.


100-300 dollars covered. Nothing from nvidia is here to combat that aside from stale maxwell parts. I distinctly remember the bile throw towards amd when the 390 series and those terrible rebrands, but at least those rebrands gave a boost in performance being souped up pre overclocked variants, NO one in good conscience would recommend some maxwell part below 300 dollars over the new rx series, even razor is not enough of a sadist and nvidia cheerleader to go THAT far.


perf/dollar goes to amd
and EVEN IF the per/dollar of the 1070 matches/comes close/is slightly ahead of the 480 series, it will do so in a higher price tier where many gamers would choose something at a lower cost to get the same job done.

 

[FSRedWing]

We are talking about the prospect of the rx480 as the performance part if Pascal wasn't around or Pascal's performance was a bit lower that it is. I should have added this.

1) so by forgoing the performance segment and taking in less cash so the competition can beat you again, and again for another year in the very segment that has highest volume sales and most margins is a good thing when you look at long term profitability when you can't gain marketshare?

2) You don't think AMD can produce enough stock of rx480?

3) Vega is inconsequential to this argument because if P10 was low end performance, Vega is high end performance and enthusiast. (pretty much going after the ti versions, and Titan)

1) AMD has been gaining market share in recent quarters without having better parts in the performance segment. Granted, we have no idea of whether or not that will continue but AMD's current strategy indicates that either they believe so or they have no other choice.

2) I don't know. But it's reasonable to assume that neither company is able to produce as much stock as they'd like right now, given the new manufacturing processes.

3) For you it might be inconsequential but for a company, it isn't. They will take into account the percentage of users willing to upgrade from current to future products and price plays at least as important a factor as performance in that calculation, especially if products are released in short order. You are more likely to buy a new product if you feel that you got your money's worth for the current one. A product you paid $199 for is more likely to be replaced shortly afterwards than a $349 one.

In the end, the answer to your original question is simply not as straightforward as it would be for you and me, if we had the choice to make either $100 million or $200 million. But for a company, as long as it can serve its debts, further losses can be acceptable if that's what it takes to turn the company around.

 

[Deleted member 93354]

Looks like amd has enough stock and lot to go around. Seriously why the fuck one would want something otherwise. If amd can create competition that's good for all of us. This crap about oh it's a failed product cuz it was suppose to be high and bla bla, it's anyone's guess. But there is no issue with performance for he price they are asking. .

Let me ask you a question. Three years from now, will a RX480 be adequate for full bells and whistles 1080p AND VR? Three years ago, the 7970 was all you needed. Now it's middle of the road performance. And I'm sure VR is going to take off (and so will the requirements)

Like I said multiple times, they are offering a cheap price on middle of the road performance. Problem is a lot of people already have middle of the road performance. It's been around for 4+ years. So what incentive is there for them to upgrade if not building a new system?

I also said AMD has a serious image problem. Fastest is usually equated with technology leadership. If they can't make the fastest, then they are lagging in technology. That's a hard image to fight.

Do I think the RX480 is a good buy for the money? Yes
Do I think it will dominate the price bracket? Yes
Do I think they will sell a boatload of them? I'm not so sure. New PC sales are down. And that performance segment has already been saturated for years.
Will NVIDIA cut their throat when the 1060 comes out? Hard to say. But it's not looking good.

 

[linkgoron]

Well, when you read this comment, and couple that with the fact that Raja in the interview with VentureBeat mentioned Polaris 10 being a "premium graphics" part, is it that hard to make the logical conclusion? Of course AMD had to be vague, they'd be stupid to outright say whatever, because at that point they had no idea what Pascal was capable of.

They didn't announce Vega back then, so is it that hard to understand that that comment pertains to Polaris?

No, it is not hard. However, it is a mistake. Polaris is not mentioned in the slide or in the quote.Do we even know if Polaris has a memory controller which supports HBM? Regarding "premium graphics" - they're still calling it "VR premium" whatever that marketing speak means.

Middle of 2016 obviously meant Polaris. This part meant that they planned on beating existing GPUs from NVIDIA, such as Titan X and GTX 980Ti.

We'll wait for Polaris benchmarks, but again - I cannot believe that anyone at AMD believed that P10 would beat Titan X or anything similar. They're talking about Finfet etc, which includes Vega, and that's what they're talking about. Polaris is just way too small to compete with the 980ti or Fury X. He states multiple times that they're going for perf/watt, and not going for absolute performance. Even if they planned on having Polaris 10 at Fury (vanilla) level, it still isn't even close to any decent AIB 980ti.

And no, nobody said Polaris 11 was supposed to be the 480 line. Entry desktop for AMD has always been x70 and below. But it's not implausible that Polaris 10 was supposed to have at least 2 cuts, 480 and 490, is it?

It's not implausible, however it is not likely and I do not believe that it is the case. When was the last time we saw one "die" fill 4 cards (non-OEM/rebrands)? GK110? and it took them around a year to release all of them. Also, where in the performance stack would that put Vega 10 and Vega 11?

I've seen people try to understand how a bigger and improved Polaris (i.e. Vega) would perform, but we've seen that performance doesn't always scale linearly. e.g. 980 vs 960, 390x vs 380x or AIB 980ti vs 980 (note that the 980ti is not even a full chip). Of note, the 390x is only 22% bigger than the 380x but is 55% faster according to TPU.

I do think, however, that AMD thought that they'd release Polaris before Pascal. In pre-Pascal days Polaris 10 could be considered "performance" and not mainstream.

 

[razor1]

1) AMD has been gaining market share in recent quarters without having better parts in the performance segment. Granted, we have no idea of whether or not that will continue but AMD's current strategy indicates that either they believe so or they have no other choice.

if you think 4% in 3 quarters is a good then man I don't know what to say. 4% in 3 quarters is what, 1.3% per quarter, yeah that isn't because of good products. That is because they have products in channel vs not having products in channel, but those products in channel are still not very attractive.

2) I don't know. But it's reasonable to assume that neither company is able to produce as much stock as they'd like right now, given the new manufacturing processes.

Its not amount of sellable products end users can buy at the moment, its in the entirity of the life time of the product we are talking bout and when certain things are being launched.

3) For you it might be inconsequential but for a company, it isn't. They will take into account the percentage of users willing to upgrade from current to future products and price plays at least as important a factor as performance in that calculation, especially if products are released in short order. You are more likely to buy a new product if you feel that you got your money's worth for the current one. A product you paid $199 for is more likely to be replaced shortly afterwards than a $349 one.

For this discussion it is, because if the rx480 ended up as a performance part even at the low end and the low end performance part is the one that holds majority % of marketshare, Vega even with it being 6 months out, wouldn't make a difference. Because AMD would be making hand over fist fulls of money vs going to lower tiers. They have been relegated to doing a midrange part because Pascal's performance couldn't be gotten with Polaris.

Do you know the mentality of a person that buys a 100 buck card vs a person that buys a 200 buck card vs the guy that buys a 300 buck card vs a person that buys a 500 buck card, vs the ultra high enthusiast who doesn't care about the price?

everyone outside of the ultra enthusiasts don't upgrade cards quickly, and as the tiers go down the upgrade cycle time usually drops too.

In the end, the answer to your original question is simply not as straightforward as it would be for you and me, if we had the choice to make either $100 million or $200 million. But for a company, as long as it can serve its debts, further losses can be acceptable if that's what it takes to turn the company around.

It is very straight forward, by skipping the performance market for 6 months, AMD would lose more than gain. And AMD can't afford to be in any position of loss. In their eyes they missed an opportunity because we have three options

• They misread where Pascal performance parts are going to be
• They just thought they could ride out the wave with midrange parts till their performance parts were ready.
• Cause and effect once Pascal came out they switched strategies cause Polaris wasn't in the same league

All three of these mean the same thing, they loose the performance market for a year.

So you think they would have gone without a single thought about the performance market, I think that is not realistic at all.

One of those three things stopped AMD from perusing the performance market.

The first one can be correlated to the other two. As nV was surprised at the clocks they were getting from Pascal, their target goals of performance were higher than expected.

Once we start seeing the overclocking of these card, if indeed it can go up to 1400 with stock voltages, thats another 10% increase in performance if linear, that is a good when you think about 150 watts at nano level performance. If they could bump that up to 180 watts, you get FuryX level performance maybe, that is a higher frequency and voltage though probably the 1600 mhz? If pascal wasn't out, would that have looked good for a performance card

180 watts at FuryX, 980ti, TitanX performance?

That would be a pretty good entry level performance card if there was no 1070.

 

[Ieldra]

We are talking about the prospect of the rx480 as the performance part if Pascal wasn't around or Pascal's performance was a bit lower that it is. I should have added this.

1) so by forgoing the performance segment and taking in less cash so the competition can beat you again, and again for another year in the very segment that has highest volume sales and most margins is a good thing when you look at long term profitability when you can't gain marketshare?

2) You don't think AMD can produce enough stock of rx480?

3) Vega is inconsequential to this argument because if P10 was low end performance, Vega is high end performance and enthusiast. (pretty much going after the ti versions, and Titan)

There's two Vega gpus planned though

It really is a case of tesla model s class drivers looking at the pending model 3 with turned up noses. What does THAT do for ME ?!??? And as we all know

Sorry for quoting you guys, I'm on mobile and it won't won't go away



Ashes of the Singularity

Extreme 1080p. Mfw '390x' performance D:

Your math is wrong! They said
It was 51% gpu scaling, they said.

You don't understand, they said

 

[NKD]

Let me ask you a question. Three years from now, will a RX480 be adequate for full bells and whistles 1080p AND VR? Three years ago, the 7970 was all you needed. Now it's middle of the road performance. And I'm sure VR is going to take off (and so will the requirements)

Like I said multiple times, they are offering a cheap price on middle of the road performance. Problem is a lot of people already have middle of the road performance. It's been around for 4+ years. So what incentive is there for them to upgrade if not building a new system?

I also said AMD has a serious image problem. Fastest is usually equated with technology leadership. If they can't make the fastest, then they are lagging in technology. That's a hard image to fight.

Do I think the RX480 is a good buy for the money? Yes
Do I think it will dominate the price bracket? Yes
Do I think they will sell a boatload of them? I'm not so sure. New PC sales are down. And that performance segment has already been saturated for years.
Will NVIDIA cut their throat when the 1060 comes out? Hard to say. But it's not looking good.

Lot of people already have 390x/gtx980 performance? I really don't think so. I know I don't. I would rather spend 250 and have be obsolete in 2 years then spend 400+. So everyone doesn't have that performance and now they have a chance to get it. You can agree to disagree but when you say middle of the road performance when the same performance right now will cost you 350+ unless you are willing to spend 450 on 1070. Price/performance 229 getting your performance of 390x/Gtx980 those are 350 right now. Plus more potential if it overclocks well. That's all I am saying. We don't always have to talk down a product just because competition has higher ends available. I always look at cost/performance and OC ability. If it hits those three marks. I don't care about power. That's a fanboy thing. 20-30 more watt isn't gonna killed me.

 

[bildad]

Ya'll are missing something. The price is nice and all, but I don't think they're going to sell as many as you think. The 480 has a problem, and that is it's a mid range card that offers nothing for those already in the mid range and certainly nothing for those in the high range. If you already have a 290x/390/390x then 480 offers no upgrade for you, let alone Fury. There is no point to buy one. Same goes for if you're in the current NV mid range or above. The only people this card is going to appeal to, or offer anything to, are those in the low range looking to get into the mid range for cheap or those who have no card at all. But how many of those people even know or care?

 

[TheLAWNoob]

Ya'll are missing something. The price is nice and all, but I don't think they're going to sell as many as you think. The 480 has a problem, and that is it's a mid range card that offers nothing for those already in the mid range and certainly nothing for those in the high range. If you already have a 290x/390/390x then 480 offers no upgrade for you, let alone Fury. There is no point to buy one. Same goes for if you're in the current NV mid range or above. The only people this card is going to appeal to, or offer anything to, are those in the low range looking to get into the mid range for cheap or those who have no card at all. But how many of those people even know or care?

Miner will snatch them up for sure. I'm going for about 5 to see how it goes, and my friend is getting 1 to be conservative.

Half the TDP of R9 290 is very tempting.

If worst comes to worst, I can always sell it and buy used R9 290s instead.

 

[MTDEW]

Miner will snatch them up for sure. I'm going for about 5 to see how it goes, and my friend is getting 1 to be conservative.

Half the TDP of R9 290 is very tempting.

If worst comes to worst, I can always sell it and buy used R9 290s instead.

Possible Ethereum Hashrate for the New AMD Radeon RX 480
Possible Ethereum Hashrate for the New AMD Radeon RX 480 - Crypto Mining Blog

We have received a tip from a reader claiming he was able to run an Ethereum hashrate test on the upcoming AMD Radeon RX 480 GPUs providing us with a photo proof that we are publishing here. Do note that for the moment we cannot confirm this result for sure, but it does seems legit enough for us to consider it as a highly possible. Unfortunately the hashrate that is apparently achieved at stock frequencies is a bit disappointing at just around 24 MHS for Ethereum Dagger-Hashimoto mining with about 100W, we are also told that with a memory overclock 26-28 MHS are possible with about 120W of power usage. If these results turn out to be true the Radeon RX 480 would not be that great for mining Ethereum as we suspected recently. On the other hand the RX 470 could end up being a much more interesting choice for low power Ethereum mining if it manages to achieve the same hashrate, but at a better price when it hits the markets sometime next month. Guess we’ll have to wait a bit more to see confirmations if this is the actual hashrate for RX 480 mining Ethereum, but with the 256-bit memory bus these results unfortunately are on track with the pessimistic expectations we already had instead of the optimistic version…

 

[linkgoron]

The most common argument I see is the 'amd offers unmatched performance per dollar', does it?

$240 on the 8GB 480 vs $380 on the 1070.

If the 480 offers less than 60% the performance of the 1070 it will actually offer less performance per dollar, leaving AotS data aside, the firestrike leaks show the 1070% being 50% faster if I'm not mistaken

The 480 are not for sale yet, but the latest rumors put the MSRP at 230$, not 240$.

Also, can you give me a link to any 1070 which is under 400$?
According to NowInStock the cheapest 1070 is currently 420$.

 

[Pieter3dnow]

Possible Ethereum Hashrate for the New AMD Radeon RX 480
Possible Ethereum Hashrate for the New AMD Radeon RX 480 - Crypto Mining Blog

Damn those evil miners with their evil schemes . So no miner conspiracy or is this just one hell of a jealous miner throwing of the other miners so he can keep all the cards to himself ?

 

[activeboot]

Ya'll are missing something. The price is nice and all, but I don't think they're going to sell as many as you think. The 480 has a problem, and that is it's a mid range card that offers nothing for those already in the mid range and certainly nothing for those in the high range. If you already have a 290x/390/390x then 480 offers no upgrade for you, let alone Fury. There is no point to buy one. Same goes for if you're in the current NV mid range or above. The only people this card is going to appeal to, or offer anything to, are those in the low range looking to get into the mid range for cheap or those who have no card at all. But how many of those people even know or care?

Actually what this card offers as being a mid-range card is previous generation high-end card performance for the cost of a mid-range card. As one article said "Ferrari performance for a price of a Honda".

 

[tybert7]

Ya'll are missing something. The price is nice and all, but I don't think they're going to sell as many as you think. The 480 has a problem, and that is it's a mid range card that offers nothing for those already in the mid range and certainly nothing for those in the high range. If you already have a 290x/390/390x then 480 offers no upgrade for you, let alone Fury. There is no point to buy one. Same goes for if you're in the current NV mid range or above. The only people this card is going to appeal to, or offer anything to, are those in the low range looking to get into the mid range for cheap or those who have no card at all. But how many of those people even know or care?

if over 80% of the market of discrete gpu buyers is sub 300 dollars, that means that until just recently, everything at around a 970/390 performance level on up was unavailable to them new. The 970s just barely starting selling for 299 (no point going there now over a 480), and some used 290s/390s might be below 300, but for the non used brand new gpu market, where the majority of actual buyers are, they DON'T have rx 480 performance.

Most people on these forums probably do, but step outside of yourself. Look at these charts on amazon


Amazon Best Sellers: Best Computer Graphics Cards


thousands of reviews for lowly 960 parts, and 750tis

These people are not currently getting 390/970 or higher performance, they are in the gutter. And nvidia is sure as hell not offering them a lifeline.


The CHEAPEST going rate of the 1070 right now on amazon is the freaking FOUNDERS edition at 450. I was listening to the latest jay2cents talk with the barnacles/nvidia fanboy, and he was talking about the usage of a stock performing founders edition inflating prices of so called "msrp" cards because why bother selling a faster version of a card BELOW a stock performing founders edition?

Some of the price inflation is just increased early pent up demand, but after that? How often are people going to see vanilla 1070s actually retail for 379? Especially the higher clocked/cooled variations being FASTER than a founders edition... but they are going to charge closer to msrp vs the higher numbers?


The entire founders edition existence inflates the retail prices of nvidia cards. None of this matters to the 980ti crowd, big bank hanks, dropping thousands and thousands on gpus and titans like it's nothing. But for people with actual budgets? This is an opening for a solid performing 480 to swoop in and buy a decked out kia optima/ chevy malibu instead of bumping up to some bmw 3 series with inflated pricing.

Not as fast? Can't corner as well? True, but it's less than half the going rate and does everything most people need to to get around town in style with great performance.

This is not a disaster in the making for amd.

 

[Deleted member 93354]

Damn those evil miners with their evil schemes . So no miner conspiracy or is this just one hell of a jealous miner throwing of the other miners so he can keep all the cards to himself ?

I thought they all become irrelevant cost wise as soon as FPGA/ASIC cards dedicated to mining came out.

Due to the rising hashrate of the bitcoin network caused by the introduction of ASICs to the market, GPU mining Bitcoins has become impracticable. The hashrate of most GPU units is below 1GH/s, and as of 2014, some single ASIC units are able to reach speeds of over 1,000GH/s while consuming far less power than used by a GPU. The information in this table is preserved for historical interest, but does not include many GPUs which were released after the advent of ASIC mining.

Any way, 20MH/s is pretty crappy considering you can pull 700MH/s out of a 7970. (Although not efficiently)

 

[bildad]

Actually what this card offers as being a mid-range card is previous generation high-end card performance for the cost of a mid-range card. As one article said "Ferrari performance for a price of a Honda".

From the data we've seen so far it doesn't offer previous gen high end performance though. It offers previous gens mid range performance at a low range price. To use your analogy it's a Honda for the price of a Kia. But for those who already have a Honda it doesn't offer anything new.

 

[Mav451]

Moyen just posted a rather bullish column for AMD about 30 minutes ago.
He quotes the current Amazon listing prices ($449, $849 respectively) for Pascal listing. I can sympathize with viewpoint as after all - the cards are sold out everywhere, and the few available listings are by opportunistic resellers.

His focus is primarily on the prospect that marketshare and GPUs specifically will inspire investor support.
I would disagree, as I still believe the Chinese consortium news to still be the biggest driver for investor confidence - and not WCCFTech hype articles.

Moyen references WCCFTech material continuously in his bullish column.
Now the odd part is he references enthusiast class growth ($299+), doubling from 2.9M in 2014 to 5.9M in 2015).
Why reference growth in the enthusiast sector, when Polaris is targeting $100-$300 range

He then moves onto his point about volume over ASP:

The volume-based advantage in selling cheap video cards will allow AMD to make a decent gross margin on entry-level/mid-range GPUs.

He also talks about design wins (e.g. Apple, consoles, OEMs) - nice to have sure, but not sources of disruptive growth that would actually support a true turnaround thesis.

The biggest gaffe of all in his column however?

I am again endorsing AMD as a BUY. Lisa Su is shrewd. She is applying the usual Chinese tactic of flooding the market with cheaper alternatives towards beating (the more expensive) rival brand.

Lisa Su was born in Tainan, Taiwan. Moyen associating her with the Mainland shows how little research Moyen puts into his columns

 

[harmattan]

The biggest gaffe of all in his column however?


Lisa Su was born in Tainan, Taiwan. Moyen associating her with the Mainland shows how little research Moyen puts into his columns

Moyen made just a bit of a gaffe there. It's a bit like saying Steve Jobs was using that "usual canadian strategy of being overly-controlling". Heck, I'd also be offended if I was Chinese as well for the implication the Chinese can't make quality goods.

 

[bildad]

if over 80% of the market of discrete gpu buyers is sub 300 dollars, that means that until just recently, everything at around a 970/390 performance level on up was unavailable to them new. The 970s just barely starting selling for 299 (no point going there now over a 480), and some used 290s/390s might be below 300, but for the non used brand new gpu market, where the majority of actual buyers are, they DON'T have rx 480 performance.

Most people on these forums probably do, but step outside of yourself. Look at these charts on amazon


Amazon Best Sellers: Best Computer Graphics Cards


thousands of reviews for lowly 960 parts, and 750tis

These people are not currently getting 390/970 or higher performance, they are in the gutter. And nvidia is sure as hell not offering them a lifeline.


The CHEAPEST going rate of the 1070 right now on amazon is the freaking FOUNDERS edition at 450. I was listening to the latest jay2cents talk with the barnacles/nvidia fanboy, and he was talking about the usage of a stock performing founders edition inflating prices of so called "msrp" cards because why bother selling a faster version of a card BELOW a stock performing founders edition?

Some of the price inflation is just increased early pent up demand, but after that? How often are people going to see vanilla 1070s actually retail for 379? Especially the higher clocked/cooled variations being FASTER than a founders edition... but they are going to charge closer to msrp vs the higher numbers?


The entire founders edition existence inflates the retail prices of nvidia cards. None of this matters to the 980ti crowd, big bank hanks, dropping thousands and thousands on gpus and titans like it's nothing. But for people with actual budgets? This is an opening for a solid performing 480 to swoop in and buy a decked out kia optima/ chevy malibu instead of bumping up to some bmw 3 series with inflated pricing.

Not as fast? Can't corner as well? True, but it's less than half the going rate and does everything most people need to to get around town in style with great performance.

This is not a disaster in the making for amd.

The whole FE thing has been a disaster for the consumers and market pricing. Stupid, stupid, stupid. Maybe great for NV, but screwed everyone else. I don't think we'll see a $380 1070 for a long time. If NV can get away with selling them for $450 then there's no reason for the AIBs not to do the same or higher, especially since they have even better cooling than the FE.

The point I was making is what is really the market for the 480? It can't be marketed to the mid range because if you have last gens mid range you already have the same performance so how cheap it is becomes irrelevant. You're not going to see many people spending another $200+ just to be exactly where they already are. Unless it's free or you can sell your old mid range card for $200 and get the 480 for $200, but then is it even worth the effort? That leaves the 480 to be marketed to the low end buyer. The question is how many of them will care? What percentage of that market can the 480 capture? Some percentage will be those who couldn't afford a 970/390 but wanted the performance and to them it will be appealing. But some percentage are not big gamers or even know what FPS is so how many of them can be attracted? Those running IGP mostly likely couldn't care less or have a need for any discrete graphics so I would imagine no one is going to capture much of that market.

The FE pricing fiasco aside, the 1070 does offer an upgrade path for the mid range buyers, who are generally well informed and looking for something to play the latest games. If they could afford a 970/390 then they can probably afford a 1070 and that does offer a substantial improvement and upgrade path.

 

[TheLAWNoob]

Possible Ethereum Hashrate for the New AMD Radeon RX 480
Possible Ethereum Hashrate for the New AMD Radeon RX 480 - Crypto Mining Blog

Just gotta burst my bubble don't you?

I hate you

 

[Ieldra]

The whole FE thing has been a disaster for the consumers and market pricing. Stupid, stupid, stupid. Maybe great for NV, but screwed everyone else. I don't think we'll see a $380 1070 for a long time. If NV can get away with selling them for $450 then there's no reason for the AIBs not to do the same or higher, especially since they have even better cooling than the FE.

The point I was making is what is really the market for the 480? It can't be marketed to the mid range because if you have last gens mid range you already have the same performance so how cheap it is becomes irrelevant. You're not going to see many people spending another $200+ just to be exactly where they already are. Unless it's free or you can sell your old mid range card for $200 and get the 480 for $200, but then is it even worth the effort? That leaves the 480 to be marketed to the low end buyer. The question is how many of them will care? What percentage of that market can the 480 capture? Some percentage will be those who couldn't afford a 970/390 but wanted the performance and to them it will be appealing. But some percentage are not big gamers or even know what FPS is so how many of them can be attracted? Those running IGP mostly likely couldn't care less or have a need for any discrete graphics so I would imagine no one is going to capture much of that market.

The FE pricing fiasco aside, the 1070 does offer an upgrade path for the mid range buyers, who are generally well informed and looking for something to play the latest games. If they could afford a 970/390 then they can probably afford a 1070 and that does offer a substantial improvement and upgrade path.

Man that's what people said about the 1080. In europe price is ~790 euros for 1080 FE. Not only have i seen ~700 euro FE in stock , there are 1080s starting ~620 euros but they are the blower AIB models, there's no many of them yet, You can buy a G1 gaming for 700 euros.

700 euros - VAT - ~35 euro other shit = 540 euro; 610 USD.

Gigabyte GV-N1080G1 GeForce® GTX 1080 Gaming G1 8GB GDDR5X 2560 Core VR Ready - GV-N1080G1 - DB Gamer

available now

These things sell out very fast, high demand and limited supply

gtx 1070 only FE is available @ 470 euro

 

[ChosenUno]

No, it is not hard. However, it is a mistake. Polaris is not mentioned in the slide or in the quote.Do we even know if Polaris has a memory controller which supports HBM? Regarding "premium graphics" - they're still calling it "VR premium" whatever that marketing speak means.

We don't know 100%, however it's been rumoured that Polaris has a memory controller capable of using both.

We'll wait for Polaris benchmarks, but again - I cannot believe that anyone at AMD believed that P10 would beat Titan X or anything similar. They're talking about Finfet etc, which includes Vega, and that's what they're talking about. Polaris is just way too small to compete with the 980ti or Fury X. He states multiple times that they're going for perf/watt, and not going for absolute performance. Even if they planned on having Polaris 10 at Fury (vanilla) level, it still isn't even close to any decent AIB 980ti.

They've never publicly talked about anything Vega-related, and it was an interview during Polaris launch event. Why would they talk about Vega then? Even if they did, why not explicitly mention that they were? Because by not being explicit, they opened up the interpretation that Kyle presented regarding expected vs actual Polaris performance.

It's not unreasonable for them to have designed Polaris to be around Fury X levels when mildly overclocked, Nano levels stock.

A RX490 with Nano performance stock, >= Fury X performance when overclocked, at around 130W. That, to me, sounds about right for a node jump + FinFET + architectural changes.

It's not implausible, however it is not likely and I do not believe that it is the case. When was the last time we saw one "die" fill 4 cards (non-OEM/rebrands)? GK110? and it took them around a year to release all of them. Also, where in the performance stack would that put Vega 10 and Vega 11?

I've seen people try to understand how a bigger and improved Polaris (i.e. Vega) would perform, but we've seen that performance doesn't always scale linearly. e.g. 980 vs 960, 390x vs 380x or AIB 980ti vs 980 (note that the 980ti is not even a full chip). Of note, the 390x is only 22% bigger than the 380x but is 55% faster according to TPU.

I do think, however, that AMD thought that they'd release Polaris before Pascal. In pre-Pascal days Polaris 10 could be considered "performance" and not mainstream.

If Polaris 10 had Nano performance stock, Fury X overclocked, it would place full Polaris 10 squarely in the enthusiast tier. And with a cut down die for the 480 for the mainstream segment.

Now if AMD assumed 1080 to be = 980Ti, and 1070 to be = 980, then full Polaris 10 at Nano/Fury X level stock/OCed and cut Polaris 10 at Fury/Nano levels would've made pretty much every single gamer in the world drool, especially if they priced it at, say, $500 and $400.

 

[Ieldra]

RX 480 firestrike ultra

I scored 2 977 in Fire Strike Ultra