• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

AMD Radeon Pro Vega Frontier Edition TDP and Pricing Revealed

Last time I checked getting a GPU under water would cost you WAY less than $600 ... still something doesn't add up. Also if Apple wants to stick these into their nice efficient Macs... well .... you get the picture. Furthermore it's like the article says if this is aimed at the prosumer market the gaming ones should be more aggressive with clockspeeds and whatnot, let's just hope these things deliver, if they prove to be slower than nVidia's offerings RTG has some serious trouble.
AMD is probably binning higher clocking chips for water cooling.
 
am I the only one who doesn't care about power used?

I mean, a 100 watt difference only means like $8.76 a year more if it's working 4 hours a day at 100%, which realistically isn't near that. (that's for what I pay for electricity).

someone who pays $0.25 per kWh will only be like $36 per year for 100 watts. (3$ per month).

If you can afford the card you can afford that increase (IMO).

For rendering, at 20/7 usage, it would be $180 per year, if you're using it for income, $180 per year is only $15 per month. easily justifiable.
 
Its really the heat and noise with that power that is most peoples concern. Id like to see a 120mm cool 300W+ loads...

Oh wait..amd did that already with 295x2...and that rad got HOT!!!!
 
Its really the heat and noise with that power that is most peoples concern. Id like to see a 120mm cool 300W+ loads...

Oh wait..amd did that already with 295x2...and that rad got HOT!!!!

Fury X too. Don't forget though, this has a blower too.

But cooling method is irrelevant really, at 100% the card will put out the same heat air cooled or watercooled. It's just the efficiency that changes.
 
am I the only one who doesn't care about power used?

I mean, a 100 watt difference only means like $8.76 a year more if it's working 4 hours a day at 100%, which realistically isn't near that. (that's for what I pay for electricity).

someone who pays $0.25 per kWh will only be like $36 per year for 100 watts. (3$ per month).

If you can afford the card you can afford that increase (IMO).

For rendering, at 20/7 usage, it would be $180 per year, if you're using it for income, $180 per year is only $15 per month. easily justifiable.
The power bill is not the problem. It's the heat, noise and overclockability.
 
The power bill is not the problem. It's the heat, noise and overclockability.

So extra heat equivalent to three 60 watt lightbulbs is a concern? I just have a hard time seeing it. I mean I understand it does make a difference in a room, but there are small changes you can make to reduce total heat produced in a room to offset this.

(I have a 290X lightning that pulls 450W ALONE from the wall (680 total power draw from the wall with a 7700K), I know all about heat xD.
 
am I the only one who doesn't care about power used?

I mean, a 100 watt difference only means like $8.76 a year more if it's working 4 hours a day at 100%, which realistically isn't near that. (that's for what I pay for electricity).

someone who pays $0.25 per kWh will only be like $36 per year for 100 watts. (3$ per month).

If you can afford the card you can afford that increase (IMO).

For rendering, at 20/7 usage, it would be $180 per year, if you're using it for income, $180 per year is only $15 per month. easily justifiable.
What about noise? Dissipating 100W more, will always be noisier.
 
Mentioned at least twice already. :)



And derp the heatloadload doesnt change with air or water... you really think that was the point there duality????
 
What about noise? Dissipating 100W more, will always be noisier.

Get a better fan. a full cover waterblock on a 360mm radiator for this card will not be noisier than stock, even with 100w more.

Get a lower ambient, will improving cooling for the card, without being noisier.

Allow your card to be hotter than your target temp, this will increase the delta, thus enable it to dissipate more heat with the same fan speed, won't be noisier either.

Mentioned at least twice already. :)

And derp the heatloadload doesnt change with air or water... you really think that was the point there duality????

Nope, and I hope you didn't think that's what I was saying. That's why i said efficiency. Watercooling dissipates the same heat at air cooled, but more efficiently, meaning faster and better.
 
Oh, so spend $250 more is the answer?? Gotcha.

Get a lower ambient???!!! Yeah, i want to lower my AC 2C to make a 2c difference...costs keep adding up!!!
 
Oh, so spend $250 more is the answer?? Gotcha.

He said "Always" I was just confirming that it wasn't the case applying a blanket statement as he did. No need to be patrionizing, we're having a discussion.
 
Sorry... the ideas need grounded.

You say it doesnt cost much more (power wise and that small nugget is true), yet here we are 'discussing' $250+ solutions and lowering ambient to mitigate the noise...
 
He said "Always" I was just confirming that it wasn't the case applying a blanket statement as he did. No need to be patrionizing, we're having a discussion.
And I was right, too. Because I can apply any of the solutions you mentioned to a card that uses 100W less power, and the end result will be quieter.
 
The fail is strong with VEGA. Do some homework youself and compare FE VEGA to Quadro cards, you will understand why AMD put it against TitanXp in professional benchmarks.

Too expensive to produce for some pathetic performance. VEGA is DOA. Move along people. Hype train next stop: VEGA20
 
I was answering that specifically to this.
You can answer to whatever you want, physics dictates when there's less power to dissipate, you can do it with less noise. It's really simple, I'm not even sure why we're discussing this.

Edit: Is power usage the primary factor when deciding to buy such a card? Undoubtedly no. But using more power than another product to deliver roughly the same results is definitely not an advantage. And it's also not something to be ignored, like you think it is.
 
Last edited:
So extra heat equivalent to three 60 watt lightbulbs is a concern?
I haven't seen a 60W lightbulb for years. :-D
I just have a hard time seeing it. I mean I understand it does make a difference in a room, but there are small changes you can make to reduce total heat produced in a room to offset this.
Like what? Install air-conditioning? ;-)

It's not just about the cost of electricity. It's also about the actual current needed for the building, certificates and so on.

Moreover, I can understand this is not such a great issue in sparsely populated US and Canada, but in dense european and asian countries it's a lot more important.

BTW: here you can find some figures about electricity usage per employee. Not many countries are listed, but Germany vs US is already pretty educative.
http://g20-energy-efficiency.enerda...ity-consumption-of-services-per-employee.html
 
After the major cock-up they pulled @ Computex? How can one not be a sceptic regarding this. Also charging $600 on cooling I'm sorry to say it, justifies jack-all. Furthermore how do we know that this will deliver? So far everything is in Limbo with a paper lunch on the way.
What went wromg at computex? 600 dollars is fine for people wanting a single workstation gpu and the others will likely get multiple gpu's for a xfire setup in a ultra high-end workstation. If you only need one workstation gpu, paying 600 dollars extra for improved cooling and performance may not seem too bad, especially considering even threadripper will likely be $1000+ extra including mobo for the 16-core anyway, so you end up paying relatively little extra.
 
The fail is strong with VEGA. Do some homework youself and compare FE VEGA to Quadro cards, you will understand why AMD put it against TitanXp in professional benchmarks.

Too expensive to produce for some pathetic performance. VEGA is DOA. Move along people. Hype train next stop: VEGA20
These aren't technically full-on pro cards yet. More general powerhouses aimed at basic professional use. There is a difference apparently according to nvidia and amd.

The vega pro cards for the imac pro should be more efficiënt and may have a few features making it better for certain professionals. Pro gpu's have niches as well.

Also, where is the MI25?
 
The fail is strong with VEGA. Do some homework youself and compare FE VEGA to Quadro cards, you will understand why AMD put it against TitanXp in professional benchmarks.

Too expensive to produce for some pathetic performance. VEGA is DOA. Move along people. Hype train next stop: VEGA20

I'm fairly sure you also know the prices of those quadro you mention.
 
The fail is strong with VEGA. Do some homework youself and compare FE VEGA to Quadro cards, you will understand why AMD put it against TitanXp in professional benchmarks.

Too expensive to produce for some pathetic performance. VEGA is DOA. Move along people. Hype train next stop: VEGA20

The fail is strong in your comment m8, vega FE is not a firepro card, firepro cards are the ones that are pitted against the nvidia Quadro cards... This card is for workstations...
 
Where do you see a six pin tho? Both cards seem to have 2x8pin according to those pictures. There was a a 6+8 pin card that Raja had during some presentation but afaik he said it was only eng. sample and the production cards will have 2x8pin.

I was referring to the gaming vega cards, those cards will not have 16 gb of hbm 2.0 and sure as hell not all variants of it will be 2x8 pin..
 
The fail is strong in your comment m8, vega FE is not a firepro card, firepro cards are the ones that are pitted against the nvidia Quadro cards... This card is for workstations...
Forgot firepro was a thing. Vega will be positioned above the wx7100. 8100 and 9100 maybe?
 
Back
Top