On the Quake 3 Arena w/Unichrome benches, could you move the FPS results further to the right? You've got some text overlapping there and it's quite hard for me to read what FPS was obtained.

Other than that, nice review.

Reply

I don't understand why this is "AMD's shot at SFF."

Shuttle has had AMD based SFF computers for a while now... is the processor a special SFF processor or something?

Reply

Guys, these reviews are starting to get boring.

1st - I don’t see the point of testing over clocking capabilities on "cube" PC's, it's ridiculous. The people at the market these products are aimed, this is not important, especially on such small systems, who want to do that.
2nd - I don’t see the point of buying a such small system and put a Geforce 6800 or even an ati 9800, the person that does that is not in it right mind.
3rd - the review of a system like these is more important sound quality and video quality playback as 2D display quality, and specially functionality or "extras" that systems like these have.
4th - The noise the system make while is working is also very important.
5th – And really these charts that always look the same are getting boring, or you start choosing applications/games that show “significant” differences between the various chipsets or these review get boring to read.

Reply

nserra #3 - I can't disagree with you more.

1. These systems are mainly designed as portable LAN machines for enthusiasts. And since AT is an enthusiasts, it makes sense to overclock the systems. Almost every site posts overclocking reslts, which makes complete sense.

2. Why not? Many enthusiasts want best performance from these little PCs. If you want the best gaming performance, it makes complete sense to include a 6800 Ultra and/or 9800. I think you are the one with a messed up mind. So you are saying that having an Athlon64 chip is right, but having a high-end video card is wrong? Why the heck would someone want a high-end CPU if they are not going to do gaming? Your arguments are useless.

3. Yeah, I agree with that BUT these systems are not for small businesses and schools who are not going to do anything with it. Don't forget that these systems are tragetted towards enthusiasts and nicely enough SFF proves be to a good form factor for space conscious users.

4. Yep, agreed.

5. You are wrong again. AnandTech has a selected benchmark suites which they use everytime to compare performance against competitor's products. If they were to start using new gaming and new testing methods for each product review, we, as readers, won't be able to understand the difference between each product and/or AnandTech will have to review 10 systems over and over again, which is a huge time wastage.




I give you some credit for your post but all the important points are bogus and utterly useless. Please check your facts and by the way if these reviews are getting boring...then why in the world are you even reading them. Stop complaining and do not ask AnandTech to change the testing methods upside down just because you. You are not special, in fact, you are one of the ignorant readers who doesn't know a thing about computers.

Reply

Good review. Very informative, but I have a problem.

The pictures are just fugly! I don't understand why you couple great analysis of the hardware with blurry, out of focus, poorly lit photographs.

And it's not just this review, most of the reveiws on this site have the same crappy pictures. Please...pretty please with sugar on top...send one or two of your tech gurus to a basic studio photography class at the local community college and get a camara with decent macro ability. What did you do with that Canon 10D you reviewed a few weeks ago?

This from the site that's going to be bringing you camera reviews...

Reply

#3 - You might be surprised at how some use their SFF machines. For example, the Shuttle SFF based on the 875 chipset has become a darling of the Extreme Overclocking crowd. In fact, I know 2 users who run their 875 SFF with mobile processors for low heat, the FSB at 300FSB, and top-end video. They take the machines to LAN parties.

We always publish stock performance metrics as our baseline. If you don't personally care about overclocking an SFF, then just skip that information. We understand overclocking info is not important to some, but it does matter to others.

Reply

Am I the only one that wants to know how they got NVIDIA IDE controllers on VIA chipset motherboards? Ether that, or your specs are wrong.

Reply

Firstly ... NOISE !

Even if just a subjective analysis, please include noise level tests of these SFF in the future. Some of them have been plagued with horrible levels of fan noise ...

Secondly, it would be great to have more Linux tests done on motherboards review. Even if it's just trying to install a distro and see what works/doesn't work out of the box ... mostly for onboard controllers compatibility ...

Reply

Hi, another nice review as always. But how about doing a review on Iwill's SFF dual Opteron system? According to their website, they started "sampling" the product this month, so maybe you can get your hands on one and tell us if they are worth buying.

Reply

I'm always happy to see more SFF reviews. But, based on your review, I'm pretty disappointed that Shuttle bothered to put out a new AMD SFF with such minor improvements over their previous AMD64 machine. Also, that thing is extremely ugly. What I really want to see is a Socket939 Shuttle with all of the features of their new Intel SFF (hi-quality audio, multiple drive support, cooling zones, etc.). Any word on something like that? Here's a review of the new Intel SFF.

http://www.sfftech.com/showdocs.cfm?aid=563

Reply