PowerColor Radeon RX 7900 XTX HellHound Pictured

[btarunr]

PowerColor put out its first teaser of a custom-design Radeon RX 7900 XTX graphics card, the RX 7900 XTX HellHound. The teaser is a back 3-quarter picture of the card, revealing its well-ventilated dual aluminium fin-stack heatsink and a triple fan setup, which make up a triple-slot cooling solution. The cooler's fans, and the HellHound logo on the backplate, are illuminated.

A striking aspect of this card is that it has the same set of power connectors as the AMD reference design, with just two 8-pin PCIe power connectors for a power-delivery capability of 375 W. It's possible that PowerColor is using a common board design for both the RX 7900 XTX and RX 7900 XT HellHound products. Besides the two power connectors, we spot a switch to control the LED illumination. There's another switch toward the front-end of the card, which works as a dual-BIOS selector.



View at TechPowerUp Main Site

 

[fancucker]

Igorlabs has recently uncovered a discrepancy in the way AMD reports total board power. Shunt/oscillopsy testing revealed a much higher power consumption. Please report on this.

 

[Salvo39]

Soo glad to see normal power connectors. NVIDIA subreddit mega-thread outpouring with melting 12VHPWR connectors

 

[Luminescent]

Looks too big.
When are these manufacturers realize not everybody has room for cards longer than 300mm.

 

[ZoneDymo]

Looks capable but it reminds me of another brand, anyone know what is up with HIS?

Igorlabs has recently uncovered a discrepancy in the way AMD reports total board power. Shunt/oscillopsy testing revealed a much higher power consumption. Please report on this.

Maybe link the article you are referring to?

Tpu measures power themselves so they would have found this "discrepancy".
Heck every self respecting reviewer just measures that sort of stuff themselves, GN specifically probes the actual card and never mentionend anything about this.
So yeah link the article because now it comes across very fox news esc

 

[Guwapo77]

Igorlabs has recently uncovered a discrepancy in the way AMD reports total board power. Shunt/oscillopsy testing revealed a much higher power consumption. Please report on this.

How much more can it pull, it only has two 8 pins?

 

[64K]

How much more can it pull, it only has two 8 pins?

I'm not sure what these PCIe specs really mean. I think back to the R9 295X2 review here on TPU and it had two 8pins which along with the 75 watts from the slot should have been good for 375 watts but it drew 430 watts Average Gaming and 500 watts in Peak Gaming and even drew 645 watts running Furmark.

 

[claes]

I'm not sure what these PCIe specs really mean. I think back to the R9 295X2 review here on TPU and it had two 8pins which along with the 75 watts from the slot should have been good for 375 watts but it drew 430 watts Average Gaming and 500 watts in Peak Gaming and even drew 645 watts running Furmark.

A good PSU can pull a lot more wattage than it’s rated for, as well as motherboards. This isn’t necessarily good or bad, but part and parcel with overbuilt components (which is for the better IMO). All of the major manufacturers have different formulas for “TDP,” sometimes even switching between platforms, which makes sense based on designs, but is unhelpful to the consumers, who generally aren’t following launch day beyond some the slides relating to performance increase.

Looks too big.
When are these manufacturers realize not everybody has room for cards longer than 300mm.

ATX needs to die! The future is going to lean heavily towards the mass consumer, pre-built market, where ATX has long been lost, and enthusiasts like us, where ATX makes very little sense from a power or cooling standpoint.

Mount a tower heatsink on a GPU and you’re suddenly competitive with water cooling. It’s impractical, but so are high FPI heatsink with whiny 15mm fans. There ought to be an alternative.

 

[Patriot]

Looks capable but it reminds me of another brand, anyone know what is up with HIS?


Maybe link the article you are referring to?

Tpu measures power themselves so they would have found this "discrepancy".
Heck every self respecting reviewer just measures that sort of stuff themselves, GN specifically probes the actual card and never mentionend anything about this.
So yeah link the article because now it comes across very fox news esc

Don't feed the troll. He is an intel/nvidia fancuck and brings up non-issues all the time.

 

[ratirt]

Looks capable but it reminds me of another brand, anyone know what is up with HIS?

last cards the company has released w]ere 5000 series Radeon. The company is from HongKong so maybe China related trading problems? That is why they have not released anything during that time? I did not hear anything about the company to be fair. These are just guesses.

 

[ZoneDymo]

Don't feed the troll. He is an intel/nvidia fancuck and brings up non-issues all the time.

Well that is why im asking for a source, if it really is just made up stuff (which is always odd, how can you have the opinion/positon you have when you know you have to make stuff up to support it....) then the account would be one step closer to deserving a ban.

But who knows, maybe something was discovered, not that I can find anything and again, that would go against hte likes of TPU themselves as well as GN and others.

 

[Deleted member 185088]

Looks capable but it reminds me of another brand, anyone know what is up with HIS?


Maybe link the article you are referring to?

Tpu measures power themselves so they would have found this "discrepancy".
Heck every self respecting reviewer just measures that sort of stuff themselves, GN specifically probes the actual card and never mentionend anything about this.
So yeah link the article because now it comes across very fox news esc

Nothing really important, just that AMD does things differently, basically you shouldn't use what reported by the software as tbp.
Here the calculations is made properly so nothing worth talking unless one is getting their info from the endless philistines pretending to be experts on YouTube.

Total Board Power (TBP) vs. real power consumption and the inadequacy of software tools vs. real measurements | Page 4 | igor´sLAB

Since many users and especially YouTubers as well as twitchers still take the values of tools like the MSI Afterburner or AMD’s Radeon software for the power consumption of their graphics card at face…

www.igorslab.de

 

[igralec84]

I had a 6600XT Hellhound and it was a very bad bin, probably lowest tier but quiet and good temps.

 

[ZoneDymo]

Nothing really important, just that AMD does things differently, basically you shouldn't use what reported by the software as tbp.
Here the calculations is made properly so nothing worth talking unless one is getting their info from the endless philistines pretending to be experts on YouTube.

Total Board Power (TBP) vs. real power consumption and the inadequacy of software tools vs. real measurements | Page 4 | igor´sLAB

Since many users and especially YouTubers as well as twitchers still take the values of tools like the MSI Afterburner or AMD’s Radeon software for the power consumption of their graphics card at face…

www.igorslab.de

Ah I see, thanks for the link, that is actually pretty concerning.
So this software misreads actual power consumption during game, we all know the real power consumption thanks to TPU and the like, you can look up the MSI RX6950XT Trio right now and find that during gaming it consumes 400+ watts, this was not unknown.

But for overlay software to then report way lower numbers, that is....yeah odd and concerning imo, false advertisement, something that has to be fixed.

I mean there are plenty of simple head to head videos where viewers do attach value to performance for the consumption, if the consumption is only reportedly way lower, that is just misinformation from the software, either AMD or the software manufactuer should get on that imo

 

[Vayra86]

Igorlabs has recently uncovered a discrepancy in the way AMD reports total board power. Shunt/oscillopsy testing revealed a much higher power consumption. Please report on this.

Please provide a source.

 

[azrael]

Please provide a source.

Xex360 provided a link to the article a couple of replies above. It's not really a new article, though, unless you count half a year as new. It's just been slightly reworked. No idea why. Perhaps a slow news day at Igor's Lab?

 

[nexus290]

Igorlabs has recently uncovered a discrepancy in the way AMD reports total board power. Shunt/oscillopsy testing revealed a much higher power consumption. Please report on this.

Regardless of what AMD or nVIdia say sites like this one always do their own power testing.

 

[Synthwave]

Looks too big.
When are these manufacturers realize not everybody has room for cards longer than 300mm.

This card is obviously not for everybody, even based on the price alone.

 

[doc7000]

I'm not sure what these PCIe specs really mean. I think back to the R9 295X2 review here on TPU and it had two 8pins which along with the 75 watts from the slot should have been good for 375 watts but it drew 430 watts Average Gaming and 500 watts in Peak Gaming and even drew 645 watts running Furmark.

This is because the 8 pin connector has a lot of headroom and can be capable of delivering over 300 watts of power on one power connector, however besides the R9 295X2 which is probably the only exception to this rule the 8 pin connector is treated as a 150 watt connector. So yes these cards only have 375 watts, also reviewers test power use for these cards including total system power which would 100% negate any discrepancy with how power is measured. Meaning if a test system with an RTX 4090 is using total system power of 600 watts and that same test system is using 400 watts then you know that there is a massive power use difference between those two cards.

This is why I can't stand fan boys as they will go out of there way and ignore a billion things to look for one that is an outlier and go with that extrapolating from the one outlier to billions of examples. We saw this with CPU power use between Intel and AMD, yes TDP isn't the watts that the CPU uses and yes the 105 watt TDP 5800X can pull as much as 150 watts under heavy load and yes that is way lower then what Intel CPUs pull under the same load. Yet I literally read fan boys just say (AMD lies about watts that CPU goes over 105 watts its basically the same) completely leaving out every other important piece of info that one would need to come to that conclusion.

So I can't stand fan boys as when they feel "superior" they get annoying which has been a growing issue with AMD fan boys.

 

[SLObinger]

I have a 6900XT and a 3090Ti and a kill-a-watt meter at the wall. The 3090Ti uses a hell of a lot more power. Measurement done!

 

[ZoneDymo]

I have a 6900XT and a 3090Ti and a kill-a-watt meter at the wall. The 3090Ti uses a hell of a lot more power. Measurement done!

This is known, but its more for me about this vids that show a RX6600XT going against the RTX3060 and Arc A770 and the latter two doing about 150 watts where the AMD one does 120, but apparently that cant be trusted for actual consumption

 

[RegaeRevaeb]

I'd like to see these newer cards more often pull off a less-than three-slot design. Yes, I can see greater thickness for cooling may be necessary for what's arrived, but I'd wager there's still enough of us who don't want any artificially forced choices between GPU level and PCIe slots we are allowed to keep accessible (especially for mATX blokes).

 

[Guwapo77]

I'm not sure what these PCIe specs really mean. I think back to the R9 295X2 review here on TPU and it had two 8pins which along with the 75 watts from the slot should have been good for 375 watts but it drew 430 watts Average Gaming and 500 watts in Peak Gaming and even drew 645 watts running Furmark.

I really didn't know that card pulled that much power. I went and looked at an old TPU review...it pulled up to 600W on two 8 pins. I must say, that is pretty remarkable and they still didn't melt. That card also costed $1500 and the Titan at $3000. I forgot about those ludicrous prices...