Trusted Reviews may earn an affiliate commission when you purchase through links on our site. Learn More

Sapphire Liquid Cooled Radeon X1900 XTX TOXIC Review

While I was out at CeBIT with Gordon, Sapphire was showing off a new liquid cooled Radeon X1900 XTX card and we’ve been itching to get hold of it ever since. Although some sites have reviewed engineering samples of this product, we were waiting for the finished version and it has finally arrived.

I’m not one for commenting on packaging – as far as I’m concerned it could turn up in a brown box as long it’s intact. However, Riyad was rather impressed by this packaging and was rather adamant it was mentioned. I have to admit, as packaging goes – it is pretty nice and I’m going to reluctantly suggest that it adds to the overall product experience (I’m all for style AND substance – ed.).

Included with the card is the normal Sapphire collection – component output, composite and S-Video ViVo, good quality S-Video and composite cables as well as a molex to 6-pin PCI-Express power adapter.

The card itself is ATI’s flagship X1900 XTX, which was covered here. It has 48 pixel shader processors, 16 texture units, eight vertex shaders and 16 pixel output engines. A normal XTX has a 650MHz core clock and 775MHz (1550MHz effective) memory speed, while this particular card has a slightly higher core speed of 675MHz and 800MHz (1600MHz effective) memory. That’s a minor improvement and I don’t expect it to make much difference, but we’ll see.

Clock speeds aside, what makes this card unique is its cooling solution. Instead of the reference air cooler that is so noisy noisy that I keep a set of ear plugs in my desk drawer, Sapphire has chosen to use a self-contained liquid cooling kit. As you can see, the actual cooler and the card are separate units, so it will require a spare expansion slot to accommodate this. The main graphics card has nothing but a waterblock on it, while the cooling unit contains the reservoir, pump, copper radiator and cooling fan. It is pre-filled, so you’re ready to go straight out of the box. But if you need to top up the coolant or even replace it – this is easily done.

All of the air is drawn and then removed from the case, much like the reference air cooler. This helps stop the system internals from heating up too much. On top of the cooler is a switch to run the fan at either 2,000 or 2,500rpm. From a noise perspective, I found it very difficult to tell the difference between the two so I kept it on its highest setting.

The actual cooler is manufactured by Thermaltake and is exceedingly similar to the Tide Water cooling kit, except the fan seems to be slightly larger and spins a little faster. That kit alone is around £45, so I’m surprised that the asking price of £405.38 is only around £50 more than an air-cooled version, and comes with higher clock speeds. Buying it all in one kit certainly saves a lot of fiddling around and also maintains your card’s warranty.

Some of you might be wondering how this affects CrossFire, and we’ve been told Sapphire will be releasing a limited number of CrossFire edition cards with this cooler fitted – so hopefully we’ll have an answer for you soon.

As you can see above, I was truly shocked at how quiet the cooler was. My biggest gripe with the X1900 XTX has been the cooling solution – enough to make me not recommend it on noise polution grounds alone. But this is a huge improvement. It survived our rigorous testing without falter, although I did notice the hosing was a little on the warm side. However, it was ”’still”’ cooler than the reference heatsink.

I also spent quite a while fiddling with overclocking this card. The memory was quite happy at 882MHz (1784MHz effective) which is a nice increase, while the core wasn’t so easily persuaded. At points I had the core up to 730MHz but after playing Counter-Strike: Source for even short periods it would eventually lock up at anything above the rated 675MHz. This is a little disappointing, but I wouldn’t rule out the possibility of overclocking entirely as it could well be that my particular card wasn’t that forgiving.

The nVidia cards were tested on an Asus A8N32-SLI using an Athlon 64 FX-60, 2GB of CMX1024-3500LLPRO RAM and a Seagate Barracuda ST340083A8 hard disk. Power was supplied by a Tagan 900W TG900-U95. For ATI testing, everything was kept the same except for the use of an Asus A8R32-MVP Deluxe and an Etasis 850W ET850.

The 7950 GX2 was tested using the beta 91.29 ForceWare drivers while the 7900 GT was tested using the 84.21 drivers. Both of the X1900 XTX cards were tested using the official Catalyst 6.6 drivers.

Using our proprietary automated benchmarking suite, I ran Call of Duty 2, Counter Strike: Source, Quake 4, Battlefield 2 and 3DMark 06. Bar 3DMark06, these all run using our in-house pre-recorded timedemos in the most intense sections of each game I could find. Each setting is run three times and the average is taken, for reproducible and accurate results. I ran each game test at 1,280 x 1,024, 1,600 x 1,200, 1,920 x 1,440 and 2,048 x 1,536 each at 0x FSAA with trilinear filtering, 2x FSAA with 4x AF and 4x FSAA with 8x AF.

As expected, the improvement over a standard clocked X1900 XTX was minimal, with one or two frames per second here and there. Just for comparison I threw in the 7950 GX2, to see if the Catalyst 6.6 drivers made much difference. In lower resolutions of Counter-Strike: Source the XTX cards are both slightly ahead, but as we head in to the higher resolutions the power of two GPUs shines through. This was especially evident in Quake 4, where the 7950 GX2 had leads of 30-40fps.


If you are seriously considering buying an X1900 XTX, then it is well worth paying the extra money for this card as the noise reduction is dramatic. The extra performance is just an added bonus.

However, the 7950 GX2 is simultaneously faster and quieter for the same money. The X1900 XTX on the other hand has the option of HDR and FSAA as well as the possibility of running in Crossfire (assuming you can get hold of a similarly cooled master card).

Trusted Score

Score in detail

  • Value 6
  • Features 9
  • Performance 9

Unlike other sites, we thoroughly test every product we review. We use industry standard tests in order to compare features properly. We’ll always tell you what we find. We never, ever accept money to review a product. Tell us what you think - send your emails to the Editor.