amd rx 480 is 229 dollars

jamesgalb said:
Aint that the truth.

According to nVidia fanboys, this will be a sub 970/390 card that jumps to sub 980/390x when OCed, and also jumps to 225w and super high temperatures when OCing... And there will be stock shortages...

We shall see :)
Click to expand...

NO one has stated that with any certainty, yet you want to point fingers and call out fanboys that came from a forum post of a person that stated he got his info from a brother and his brothers friends lol. Again you aren't reading everything man. Not to mention Nordic hardware even stated going past 1400 is not an easy task which contradicts what WCC article stated.. To what degree, I am assuming that voltages have to be changed quite a bit. If that is the case, it looks to be the blower is putting out around 60c temps of air, that would mean the chip is a bit hotter than the 60c? That is correct it probably will be around 80c for the GPU at stock, what would the voltage increase do to the temps?

Yeah they will increase and they are getting dangerously close to a thermal barrier

What is that thermal barrier you might ask, which you won't, because you are you, you are getting to a point where silicon goes into a what is called you guessed it (probably not) quantum tunneling which starts to happen around 115 C. This is when the resistance of silicon drops so much due to temps that it can no longer send signals properly though the chip even if you try to increase voltage more. This is when leakage starts to become uncontrollable, its all due to heat. Now you might think a "better" cooler would do the trick, Yes and NO, it might, but most likely it won't because with chips you have certain thermal hotspots, so some parts of the chip can be hotter than others, to dissipate those hotspots, more than the other parts of the chip can't happen at all times, because their is no partitioning of the cooler for thermal hotpot segments, nor can that even be done, because different bins of chips have different thermal hotspots.

Even before you get to this effect you still have to worry about electomigration which will happen as temps increase as well.
 
Last edited:
razor1 said:
On the lines of marketing from my previous post, there has been no overclockers dream screw up or akin to that yet that we have seen/heard. No skewed benchmarks by AMD that don't use AF or lower AA settings to point to, none of the usual AMD screw ups that we have see from their CPU or GPU launches as of late.
Click to expand...

Raja's running a tight ship with RTG and I like what I'm seeing. :)

well for what it can do its damn good, just like the rx480. Just can't expect the same performance at such a different price range.
Click to expand...

The corollary to this, is that if RX 480 was indeed a 980 Ti level contender, AMD would have to be retarded to price it at $200.
 
Last edited:
n=1 said:
Raja's running a tight ship with RTG and I like what I'm seeing. :)
Click to expand...

OH yeah he has been doing a good job
The corollary to this, is that if RX 480 was indeed a 980 Ti level contender, AMD would have to be retarded to price it at $200.
Click to expand...

Yeah its money they would have gotten easily, margins would have been awesome.
 
razor1 said:
NO one has stated that with any certainty, yet you want to point fingers and call out fanboys that came from a forum post of a person that stated he got his info from a brother and his brothers friends lol. Again you aren't reading everything man. Not to mention Nordic hardware even stated going past 1400 is not an easy task which contradicts what WCC article stated.. To what degree, I am assuming that voltages have to be changed quite a bit. If that is the case, it looks to be the blower is putting out around 60c temps of air, that would mean the chip is a bit hotter than the 60c? That is correct it probably will be around 80c for the GPU at stock, what would the voltage increase do to the temps?

Yeah they will increase and they are getting dangerously close to a thermal barrier

What is that thermal barrier you might ask, which you won't, because you are you, you are getting to a point where silicon goes into a what is called you guessed it (probably not) quantum tunneling which starts to happen around 115 C. This is when the resistance of silicon drops so much due to temps that it can no longer send signals properly though the chip even if you try to increase voltage more. This is when leakage starts to become uncontrollable, its all due to heat. Now you might think a "better" cooler would do the trick, Yes and NO, it might, but most likely it won't because with chips you have certain thermal hotspots, so some parts of the chip can be hotter than others, to dissipate those hotspots, more than the other parts of the chip can't happen at all times, because their is no partitioning of the cooler for thermal hotpot segments, nor can that even be done, because different bins of chips have different thermal hotspots.
Click to expand...

Okay, so we are claiming a 80c chip at stock and getting clock to hitting 115c OCed... to go with the 225w power consumption when OCed... The limited stock... the $250 starting point for 8gbs... the sub-970/390 performance... etc

I look forward to the release.
 
You haven't see the temp mesurements of videocardz have you? Nice pictures for you to easily understand lol. if you need a flip book let me know.

AMD-Radeon-RX-480-temperatura.jpg


And no I 'm saying getting close to 115 c is dangerous PS this is simple physics. And if it hits anywhere near 100 C forget about 225 watts go higher.

And when did I say sub 970/390 performance, I stated higher than those but less than 390x or 980 wtf now you are trying to put words in my mouth?

What no where to go?
 
Last edited:
jamesgalb said:
Okay, so we are claiming a 80c chip at stock and getting clock to hitting 115c OCed... to go with the 225w power consumption when OCed... The limited stock... the $250 starting point for 8gbs... the sub-970/390 performance... etc

I look forward to the release.
Click to expand...

Not sure if trolling...
 
Think it's just a reading comprehension issue.

KazeoHin said:
RTG is the best thing to happen to the radeon brand since the 9700.
Click to expand...

You mean since HD4870, because that was one Hail Mary AMD completely nailed after the disaster that was 2900 XT and the disappointment that was HD 3870.

You know come to think of it, Fury X is basically the 2015 version of 2900 XT.
 
jamesgalb said:
Lying on what? You got insulting first in prior posts in prior threads. You didnt 'respond in kind', you got upset when people said your consistency of providing negative claims using false numbers is blatant brand bashing. You started implying that other people were just stupid for not agreeing with your math.

Beyond that, the claims I listed you have made. They arent lies.
Click to expand...


Wait I even stated hypothetically and even stated someone was fibbing and that was towards that other forum post lol, Your reading comprehensions has failed you now?

You just lyed about me stating performance lower than 970, I have never stated that.

jamesgalb said:
Okay, so we are claiming a 80c chip at stock and getting clock to hitting 115c OCed... to go with the 225w power consumption when OCed... The limited stock... the $250 starting point for 8gbs... the sub-970/390 performance... etc

I look forward to the release.
Click to expand...

This wasn't you posting this.

Hmm I guess you must have a twin or your subconscious is playing tricks on you

You also quoted me saying "hypothetically" is "what I get from that forum post"

Then you exclude the fact when after I stated someone has been fibbing about overclocking results based off of Nordichardware claims.....

I can see where this is going, selective reading problems.

You got all torn up and then called me a troll because I thought that other forum post was not realistic.

That is your problem, you want that other forum post to be real?

Go for it, it will disappoint you.
 
Last edited:
Ah, you are right, you didnt make the 225w claim, you just repeated it, and you did indeed do so speculatively. That said, you used it in various posts when comparing it to the 1070 or referencing the need for PSU upgrades, so it got quite irritating to see it talked about as a meaningful number... But thats not on you on the creation of the number, so my bad there.

Still, you did get insulting first in other threads, not only with me but with others as well... And there have been a few absurd claims that I intend to take a look at again once the card is released...
 
razor1 said:
NO one has stated that with any certainty
Click to expand...
What? Tons of NV fanboys have. These forums are loaded with them.

I mean look at this awful stuff you posted right here:
razor1 said:
If that is the case, it looks to be the blower is putting out around 60c temps of air, that would mean the chip is a bit hotter than the 60c? That is correct it probably will be around 80c for the GPU at stock, what would the voltage increase do to the temps? Yeah they will increase and they are getting dangerously close to a thermal barrier What is that thermal barrier you might ask,.... you are getting to a point where silicon goes into a what is called you guessed it (probably not) quantum tunneling which starts to happen around 115 C.
Click to expand...
You jumped from 1 reasonable guess (GPU temp is probably hotter than exhaust blower temp of 60C) to insinuating that overclocking the RX480 is going to force temps of 115C which will kill the chip due to electro migration. That latter part is based on exactly zero evidence at all too. You completely made it up!! And you posted as if its practically a certainty. No mitigating statements or a "I'm just guessing help me out here guys" or even a "I'm probably wrong here but this is what I think" at all.

razor1 said:
because different bins of chips have different thermal hotspots.
Click to expand...
What?!?! Man you gotta stop making up stuff like this. A defect that might effect how a chip is binned is going to be nanoscopic in nature. Any hotspots on that feature scale don't really matter.

razor1 said:
Even before you get to this effect you still have to worry about electomigration which will happen as temps increase as well.
Click to expand...
Are you trying to make a general statement against overclocking in general or what? If you are then fine, just make sure to do a similar post on the dangers of overclocking a 1080 or 980Ti over in the Nvidia forum.

Realistically so long as you keep your temps and voltages sane the chip will become obsolete well before the extra wear and tear from overclocking kill the chip. Really its going to be voltages that most people should worry about if they're concerned about the life of their hardware.
 
mesyn191 said:
What? Tons of NV fanboys have. These forums are loaded with them.

I mean look at this awful stuff you posted right here:
You jumped from 1 reasonable guess (GPU temp is probably hotter than exhaust blower temp of 60C) to insinuating that overclocking the RX480 is going to force temps of 115C which will kill the chip due to electro migration. That latter part is based on exactly zero evidence at all too. You completely made it up!! And you posted as if its practically a certainty. No mitigating statements or a "I'm just guessing help me out here guys" or even a "I'm probably wrong here but this is what I think" at all.
Click to expand...

Have you overclocked cards before?

I have and I have done extreme overclocking with LN,

You don't understand that Finfets don't have the same tolerance as planer transistors, when it comes to thermal barriers. Its not even close.

What?!?! Man you gotta stop making up stuff like this. A defect that might effect how a chip is binned is going to be nanoscopic in nature. Any hotspots on that feature scale don't really matter.
Click to expand...

You don't have the same voltages across the entire chip, and this will change based on bins. This is why you get some bins that just don't require the same voltage or sometimes they require higher voltage. Also based on how the chip is designed, you will have many different voltages across a single chip.

We can see this with Fiji, and the nano vs the Fury X.

I am not talking about defects.

Are you trying to make a general statement against overclocking in general or what? If you are then fine, just make sure to do a similar post on the dangers of overclocking a 1080 or 980Ti over in the Nvidia forum.

Realistically so long as you keep your temps and voltages sane the chip will become obsolete well before the extra wear and tear from overclocking kill the chip. Really its going to be voltages that most people should worry about if they're concerned about the life of their hardware.
Click to expand...


I am reiterating to what Nordichardware stated, going to 1400 is no problem, going past that, things get harder, why, you need better cooling, you need more voltage, the stock stuff won't cut it anymore.
 
Last edited:
razor1 said:
Have you overclocked cards before?
Click to expand...
Yes. With air, water, and for a brief time with a single stage phase change unit.

razor1 said:
You don't understand that Finfets don't have the same tolerance as planer transistors, when it comes to termal barriers. Its not even close.
I have and I have done extreme overclocking with LN
Click to expand...
Overclocking with LN2 is a totally different thing than air, water, or even single stage phase change though. If you understand anything at all about overclocking you wouldn't make any comparisons based on it. Sub -100C temp OC'ing is generally a stunt anyways and what it does it have to do AT ALL with any thing you were claiming before?! NOTHING! Look you gotta post some hard evidence to back up what you were saying earlier otherwise I'm calling BS on it. edit: Not vague platitudes either. Every process shrink is always more susceptible to electro migration than the last and it hasn't been a big issue yet for overclocking. Nvidia's chips are being made on the same process and are people having their stuff die right and left with 10-20% overclocks? NO.

razor1 said:
You don't have the same voltages across the entire chip, and this will change based on bins.
Click to expand...
You're talking about differences of a fraction of a volt which the chip will be engineered to deal with anyways, since the engineers are aware of this problem and have to account for it in their designs to get a decent supply of good bins, and will vary not only on location but from clock to clock. IOW what you're talking about doesn't matter at all.

razor1 said:
We can see this with Fiji, and the nano vs the Fury X.
Click to expand...
Huh?? The problem with those chips is that 20nm didn't pan out at TSMC and AMD was forced to make them on 28nm. It has nothing to do with what you're talking about.

razor1 said:
I am reiterating to what Nordichardware stated, going to 1400 is no problem, going past that, things get harder, why, you need better cooling, you need more voltage, the stock stuff won't cut it anymore.
Click to expand...
No. You're adding your own insinuations about the risks of overclocking the RX480 and it reeks of FUD. If all you'd posted was those vague, and obvious, platitudes no one would care.
 
mesyn191 said:
Yes. With air, water, and for a brief time with a single stage phase change unit.


Overclocking with LN2 is a totally different thing than air, water, or even single stage phase change though. If you understand anything at all about overclocking you wouldn't make any comparisons based on it. Sub -100C temp OC'ing is generally a stunt anyways and what it does it have to do AT ALL with any thing you were claiming before?! NOTHING! Look you gotta post some hard evidence to back up what you were saying earlier otherwise I'm calling BS on it.
Click to expand...

I'm not saying they are the same they have very different techniques and you have to be much more careful with LN.

Also I also posted a pic of what the exhaust temps are for the rx480 and that is why I think some around 80 c is the chips temp at load stock.

You're talking about differences of a fraction of a volt which the chip will be engineered to deal with anyways, since the engineers are aware of this problem and have to account for it in their designs to get a decent supply of good bins, and will vary not only on location but from clock to clock. IOW what you're talking about doesn't matter at all.


Huh?? The problem with those chips is that 20nm didn't pan out at TSMC and AMD was forced to make them on 28nm. It has nothing to do with what you're talking about.
Click to expand...

Huh? IS RIGHT, look at the NANO, its the same damn chip as Fury X but it didn't need a water cooling system, it used less voltage, and when you put the fans up top it could overclock to Fury X levels, and it still used less power! Even with a hotter GPU! What does that tell you? You are talking about defects before, WTF do defects have to do with that? Yeah ok what ever. why are you bringing up 20nm and problems with that? Does that have anything to do with FIJI, do you think they planned Fiji on 20nm and fell back on 28 I don't think so at all, I think they had another chip for 20 and Fiji was a back up plan that was always on 28nm, it takes years to design chips, not the 1 and half years after 20nm failed.
No. You're adding your own insinuations about the risks of overclocking the RX480 and it reeks of FUD. If all you'd posted was those vague, and obvious, platitudes no one would care.
Click to expand...

You go back to physics class and tell me why the NANO does what it does and get back to me on this. Yeah that won't do it for ya you will need college level EE classes.
 
Last edited:
The nano is likely a binned fiji which uses less power at any given state than the fury x. It makes for an odd market position when you sell the higher binned chip as a lower slot product, but that's really the only way it makes sense. The lower binned chips needed the better cooling of the AIO and the higher ones could deal with air.

My own opinion on fiji is that it was not planned at all until after 20nm had failed and nvidia was pushing maxwell. My thinking is that it's just a bigger Hawaii with HBM instead of gddr.
 
Darakian said:
The nano is likely a binned fiji which uses less power at any given state than the fury x. It makes for an odd market position when you sell the higher binned chip as a lower slot product, but that's really the only way it makes sense. The lower binned chips needed the better cooling of the AIO and the higher ones could deal with air.

My own opinion on fiji is that it was not planned at all until after 20nm had failed and nvidia was pushing maxwell. My thinking is that it's just a bigger Hawaii with HBM instead of gddr.
Click to expand...


They won't have had to the time to do a full GPU transition from 20 to 28nm, literally, after 20nm failed, Fiji was out in a year and half if I remember correctly. Well had samples ready at least.
 
razor1 said:
I'm not saying they are the same they have very different techniques and you have to be much more careful with LN.
Click to expand...
Of course, but then why bring it up if it matters not at all to the discussion? To prove your OC chops? You could've overclocked with liquid helium before and you'd still be off base here.

razor1 said:
look at the NANO, its the same damn chip as Fury X... when you put the fans up top it could overclock to Fury X levels, and it still used less power! Even with a hotter GPU! What does that tell you?
Click to expand...
That its a better bin of the same chip that was set to lower stock clocks and volts to be sold in a smaller form factor to keep heat down. What is the surprising reveal here that I'm supposed to see? You have to prove nano/micro hot spotting is a major issue at fault here, not that some chips bin better than others.

edit: Have you posted yet over in the Nvidia forum about the extreme dangers of overclocking their 1080s and 1070s yet? You should probably do that if you're serious about this being such a disasterous issue.

razor1 said:
Pointing that better binsYou are talking about defects before, WTF do defects have to do with that?
Click to expand...
Defects effect how the chip operates which determines how the chip will be binned and sold. This is common knowledge in the industry. You don't think all those IC's come out perfect every time do you?

razor1 said:
why are you bringing up 20nm and problems with that? Does that have anything to do with FIJI, do you think they planned Fiji on 20nm and fell back on 28 I don't think so at all, I think they had another chip for 20 and Fiji was a back up plan that was always on 28nm, it takes years to design chips, not the 1 and half years after 20nm failed.
Click to expand...
Fiji is hot as hell and huge on 28nm. It clearly didn't make much sense at all and there was news that AMD was going to use 20nm and suddenly that all just vanished, NV meanwhile started complaining about TSMC's 20nm, and we got the Fiji that was shipped. Not hard to guess what happened here.

razor1 said:
They won't have had to the time to do a full GPU transition from 20 to 28nm, literally, after 20nm failed, Fiji was out in a year and half if I remember correctly. Well had samples ready at least.
Click to expand...
These days they run multiple design teams working on fall backs in case something goes wrong with a process. They probably had a 28nm fall back planned from the get go.
 
razor1 said:
They won't have had to the time to do a full GPU transition from 20 to 28nm, literally, after 20nm failed, Fiji was out in a year and half if I remember correctly. Well had samples ready at least.
Click to expand...
You may be right there. I don't know what the time lines are like. That said the HBM project has been going on for a long long time. I remember anandtech doing an article on it back when it was still ATI. As pretty much everyone noted when the fury came out; it's unbalanced. Too few ROPs for the number of shaders and a few other things (IIRC). To me it seems like a rush job in response to maxwell.
 
How about the people that dislike AMD piss off and let this thread get back on topic! You guys just keep coming up with shit to argue about.

390x-ish performance with 8gb for $229 will be pretty damn good. I paid $250can for my 280x and its looking like the 480 will dance circles around it. Better check the couch for change so I can pick one up.
 
mesyn191 said:
Of course, but then why bring it up if it matters not at all to the discussion? To prove your OC chops? You could've overclocked with liquid helium before and you'd still be off base here.
Click to expand...

I Haven't overclocked in over 10 years, just don't need to do it anymore.

That its a better bin of the same chip that was set to lower stock clocks and volts to be sold in a smaller form factor to keep heat down. What is the surprising reveal here that I'm supposed to see? You have to prove nano/micro hot spotting is a major issue at fault here, not that some chips bin better than others.
Click to expand...

You want proof,

here

Buy it

Voltage binning under process variation

Defects effect how the chip operates which determines how the chip will be binned and sold. This is common knowledge in the industry. You don't think all those IC's come out perfect every time do you?
Click to expand...

I agree I never stated anything contrary to that.

Fiji is hot as hell and huge on 28nm. It clearly didn't make much sense at all and there was news that AMD was going to use 20nm and suddenly that all just vanished, NV meanwhile started complaining about TSMC's 20nm, and we got the Fiji that was shipped. Not hard to guess what happened here.
Click to expand...


These days they run multiple design teams working on fall backs in case something goes wrong with a process. They probably had a 28nm fall back planned from the get go.
Click to expand...

yet you are saying after what ever failed on 20nm, they transitioned over to 28nm, nope, they already had the design ready for both 20 and 28nm different chips and once 20nm failed, they already had started on 28nm designs, which they picked up.
 
mesyn191 said:
That is a good paper but it has nothing to do with what you're talking about at all.
Click to expand...
Voltage binning provides an opportunity to improve yield for very strict power constraints.We analyzed this opportunity by computing optimal binning for different power requirements.
Click to expand...

You don't think that is what they did with Nano?
Process variation is recognized as a major source of parametric yield loss, which occurs because a fraction of manufactured chips do not satisfy timing or power constraints. On the other hand, both chip performance and chip leakage power depend on supply voltage. This dependence can be used for converting the fraction of too slow or too leaky chips into good ones by adjusting their supply voltage. This technique is called voltage binning [4]. All the manufactured chips are divided into groups (bins) and each group is assigned its individual supply voltage. This paper proposes a statistical technique of yield computation for different voltage binning schemes using results of statistical timing and variational power analysis. The paper formulates and solves the problem of computing optimal supply voltages for a given binning scheme.
Click to expand...

No? That sounds exactly like what they did with Nano.


Lets get back to hotspots

Chip Design » On-Chip Thermal Analysis Is Becoming Mandatory by Peter McCrorie

Hot spots occur not only on the die, but also on the package. This is due to the uneven distribution of power dissipation both on the die and in the substrate. The old approach of averaging out the power across the die and the substrate is no longer accurate enough. It’s now necessary to model power distribution across the die as well as individual structures in the package, such as substrate metal, wirebonds, bumps, and balls.
Click to expand...
 
Last edited:
Darakian said:
You may be right there. I don't know what the time lines are like. That said the HBM project has been going on for a long long time. I remember anandtech doing an article on it back when it was still ATI. As pretty much everyone noted when the fury came out; it's unbalanced. Too few ROPs for the number of shaders and a few other things (IIRC). To me it seems like a rush job in response to maxwell.
Click to expand...


I just can't see it being a rush job not that fast, I mean, yeah they had a year and half, but as a back up plan Fiji was pushing the limits of the Fab process size wise so the had to cut somethings. which actually didn't hurt them 64ROPs is enough even for 4k. I would think the delay of releasing Fiji would have come from process related issues, more than AMD's size, just because of the design of Fiji and the power usage. Not the actually fab.
 
No, lets get back to the Nano :), tempted to go CFX - Just wish the 480 came out faster to drop those prices more.
 
razor1 said:
I just can't see it being a rush job not that fast, I mean, yeah they had a year and half, but as a back up plan Fiji was pushing the limits of the Fab process size wise so the had to cut somethings. which actually didn't hurt them 64ROPs is enough even for 4k. I would think the delay of releasing Fiji would have come from process related issues, more than AMD's size, just because of the design of Fiji and the power usage. Not the actually fab.
Click to expand...
Lets continue this via a new thread or via PM.

Back on topic, is there any word on what sort of cooler the 470 is using?
 
razor1 said:
You don't think that is what they did with Nano?
Click to expand...
Of course they did some sort of binning process that was based on voltages each cheap could function at but that has NOTHING to do with what you're talking about. At all. Its just one step among many involved!!

Neither does you're other paper. Which is just talking about how hot spots are becoming an issue BUT says nothing at all how its the major factor when determining bins nor does it talk at all about hot spot effects when overclocking/overvolting.

Now unless you got some solid evidence on this issue you should probably drop it. Or instead maybe go post over in the Nvidia forum how everyone is going to kill thier 1080s and 1070s with the 10-20% overclocks they're doing. Something which you mysteriously seem not to have done still.

razor1 said:
I just can't see it being a rush job not that fast,
Click to expand...
They didn't have it complete and they needed something to compete with nvidia which is why it looks like the rush job it is. To do a proper redesign probably would've taken 2-3 years. AMD didn't have the resources then or now to finish 2, really 3 (Polaris still would've been worked on at that time), major GPU designs at the same time.

razor1 said:
I agree I never stated anything contrary to that.
Click to expand...
Yes you did.
Pointing that better binsYou are talking about defects before, WTF do defects have to do with that?
Click to expand...
 
Last edited:
Darakian said:
Back on topic, is there any word on what sort of cooler the 470 is using?
Click to expand...
None official as far as I can see. 3rd party vendors will probably put a version out with a better than stock heatsink within a few months of launch though if the default one sucks.
 
mesyn191 said:
None official as far as I can see. 3rd party vendors will probably put a version out with a better than stock heatsink within a few months of launch though if the default one sucks.
Click to expand...
I'm a weird one, but what I'm looking for is a low(er) end card with a blower cooler. I'm hoping that the 470 might get the same cooler as the 480. That would make my day.
 
Seeing as how the stock RX480 heatsink/fan is supposed to be pretty good a 470 with that unit on it would probably be awesome if you want low noise + low temps so that makes sense to me.
 
n=1 said:
The corollary to this, is that if RX 480 was indeed a 980 Ti level contender, AMD would have to be retarded to price it at $200.
Click to expand...
Maybe. Depends on (edit)what they expect Nvidia to price their 'mid-range' part at. Nvidia seems to want to drive prices up at least a bit this time around so a 1060Ti, or whatever they'll call it, for $250-300 wouldn't shock me.

I think AMD knows their brand is perceived as 2nd rate despite their best efforts so its not unreasonable that they would try and compete on price.
 
Great that the 8GB version looks like it will have an MSRP of $229. If it does perform as 390x at stock and up to Fury overclocked, that's not a bad generational upgrade (esp. since 480 is considered their mid-range), albeit not fantastic. Lower power, good overclocking (likely at least better than the insufferably-bad OC-er 390x), smaller footprint -- good stuff. That said, three things are clear to me:
  • No way in holy hell anyone is getting this card for MSRP at launch. I'd guess the lowest retailers will charge out the gate is $250-60.
  • The 8GB version will be in vastly short supply for the first month or two while AMD tries to move their 4GB version.
  • AIB versions won't likely be available until well into July
 
Last edited:
harmattan said:
Great that the 8GB version looks like it will have an MSRP of $229. If it does perform as 390x at stock and up to Fury overclocked, that's not a bad generational upgrade (esp. since 480 is considered their mid-range), albeit not fantastic. Lower power, good overclocking (likely at least better than the insufferably-bad OC-er 390x), smaller footprint -- good stuff. That said, two things are clear to me:
  • No way in holy hell anyone is getting this card for MSRP at launch. I'd guess the lowest retailers will charge out the gate is $250-60.
  • The 8GB version will be in vastly short supply for the first month or two while AMD tries to move their 4GB version.
  • AIB versions won't likely be available until well into July
Click to expand...

I'm hoping that XFX floods Best Buy with cards. I might actually make the 4 hour trip get one.
 
cageymaru said:
I'm hoping that XFX floods Best Buy with cards. I might actually make the 4 hour trip get one.
Click to expand...

Unfortunately, BB has been jumping on the "retailer tax" bandwagon of late as well. Also, I'd say 4 hours or your free time and the gas are probably worth a heck of a lot more than the premium retailers will be charging :) Either bite the bullet on paying extra or relax and wait it out a bit (which may be the way to go since prices and availability of AIB versions, both from AMD and NV, will be better in a month).
 
You must log in or register to reply here.
Back
Top