Hardware The bit-tech Hardware Awards 2008

[Bindibadgi]

http://www.bit-tech.net/hardware/200...-awards-2008/1

Earlier in the week, we published the best games of the year and today it's time to look back at what's happened in the technology industry over the last year. Find out which products are the best of the year, and also which innovations have caught our eye during 2008.

[Tejstar]

Great read, I love these end-of-year 'Best of' lists!

Surprised the Samsung SpinPoint F1 1TB won the best storage, given that there have been a number of users with problems with this particular HDD. Maybe it's down to pot luck.

Agree with the choice of graphics card and CPU (Q6600 still going very strong), although I did think a Wolfdale may win that category.

[CSMR]

Q6600, complete nonsense. This is last-generation technology. There's been a 45nm shrink since then (Q9xxx); did bit-tech miss it? As usual for a die shrink major benefits in speed and power consumption. Maybe we'll hear about it on bit-tech next year?

[teamtd11]

Originally Posted by CSMR

Q6600, complete nonsense. This is last-generation technology. There's been a 45nm shrink since then (Q9xxx); did bit-tech miss it? As usual for a die shrink major benefits in speed and power consumption. Maybe we'll hear about it on bit-tech next year?

I shall Quote Burnout and Mankz to answer this...

Originally Posted by Burnout21

i agree with mankz on this one, the jump from the Q6600 to the Q9450 just wasn't big enough in performance to justify the cost. Dare i say I7 isn't worth the cost vs performance from a Q6600 at the moment..

Great Article

[Tim S]

Originally Posted by CSMR

Q6600, complete nonsense. This is last-generation technology. There's been a 45nm shrink since then (Q9xxx); did bit-tech miss it?

No.

As usual for a die shrink major benefits in speed and power consumption.

At a price.... and what's wrong with a 3.3GHz+ Q6600? You don't need anything faster than that currently.

[Mankz]

Very good, well done Bit-Tech..

Glad to see the little ITX board getting a good mention...

[Ending Credits]

As usual for a die shrink major benefits in speed and power consumption.

So's a Caterham R500 and we don't all have those.

[naokaji]

Originally Posted by Tejstar

Surprised the Samsung SpinPoint F1 1TB won the best storage, given that there have been a number of users with problems with this particular HDD.

Most of the issues reported with the F1 drives where about the 750GB model.

Originally Posted by CSMR

Q6600, complete nonsense. This is last-generation technology. There's been a 45nm shrink since then (Q9xxx); did bit-tech miss it? As usual for a die shrink major benefits in speed and power consumption. Maybe we'll hear about it on bit-tech next year?

Q9xxx suffered from horrible avalaibility for several monthes following its launch and brutally inflated prices once they where avalaible, the 45nm dual cores where avalaible in masses here, in america it was exactly the reverse situation with the dual cores being in short supply and the quads being heavily discounted due to oversupply. Simply put, the 45nm C2D /C2Q launch was the worst porduct launch in 2008.
On top of that the there is the issue with the very low multi of the affordable version which meant the cheapest 45nm quads oc'd worse than the Q6600.

Sure, the 45nm quads where attractive in their rather short life (starting from actual avalaibility to I7 introduction), but no where as much bang for the buck as the Q6600 and no where as fast as I7.

[Bindibadgi]

Originally Posted by CSMR

Q6600, complete nonsense. This is last-generation technology. There's been a 45nm shrink since then (Q9xxx); did bit-tech miss it? As usual for a die shrink major benefits in speed and power consumption. Maybe we'll hear about it on bit-tech next year?

Actually, it was last years innovation.

[FeRaL]

Originally Posted by Bindibadgi

Actually, it was last years innovation.

Yeah but it was soooo good, it stomped everything in 2008 as well. Not a bad chip if it can get best in class for two straight years even when new stuff has come out....

[wuyanxu]

q6600 worth every penny, even more so as its price has risen back up to what i got it for :P

i am surprised that the ThermalRight Ultra 120 eXtreme has won the best cooling product, it's last years, and it's been challanged by those direct heatpipe contact coolers.

RV770 truely deserve the praise it's getting, it's well designed, cheaper to make and very competitive to nVidia's big-chip solutions.


well, q6600 and TRUE won even when they are last year's, why didn't Antec p182 won the best case?

[n3mo]

Well, Q6600 might be a little controversial, but it is the best quad-core from Intel. i7 is ridiculously expensive (which was to be expected, Intel would like it best if they could make you sell your house to buy it) and Q9xxx was hard to get and too expensive.
Personally I would put E8600 in exchange for Q6600. Most people don't need quad-core (honestly, how many people use software capable of handling multiple threads?) E8600 overclocked much higher and was generally much faster in most cases (take games for instance - E8600 is soo much faster than Q6600)

As much as I don't like Intel (and avoid their products at all times) I have to give it to them: E8600/8400 is THE gaming CPU right now.

All in all, this is a great, comprehensive article. Nicely done.

[Bindibadgi]

Originally Posted by wuyanxu

i am surprised that the ThermalRight Ultra 120 eXtreme has won the best cooling product, it's last years, and it's been challanged by those direct heatpipe contact coolers.

Direct heatpipe coolers aren't as good - the contact often leaves air gaps between the pipes and base and sometimes because they are stamped into position it damages them.

Nemo - if you say E8600, then why not E8400? It's a lot less expensive and still does 4GHz. But, by that argument, why not the E5200 then? Ours did 4GHz... The only advantage of the E8600 is for extreme overclockers, and they dont actually buy a lot of CPUs compared to the rest of us.

The truth is, there were a LOT of Q6600 sales this year even still and consistently throughout the year it's been recommended in thread after thread after thread.

[Toka]

For me the innovation of the year was the announcement by Intel that they had a viable bulk manufacturing process for the 32nm node.

Whats even more astonishing is that they are not relying on some esoteric EUVL method (this would require a completely new fab plant, maybe see it at the 22nm node) they are still using good old 193nm ArF eximers along with a high RI (refractive index) lens and some double patterning (the process of exposing the photoresist to separate patterns in order to produce features on a smaller scale).

In an industry where true cutting edge research is passed on so rapidly to the end user (think high k gates, or 45nm chips) I find the pace of progress almost frightening. We are now able to make CPUs on a scale where quantum mechanical effects are to be expected, and it wont be too long before we run up against real tunnelling losses at the 16nm node. Perhaps its time to take a moment and think about just how awesome our scale of precision manufacturing has become

[CSMR]

The Q9300 is not expensive, released at $266 according to wikipedia, available for less now. It's an impressive piece of technology, fitting into roughly the same power envelope as the 65nm dual cores, and considerable power advantages over the Q6600 (and to you skinflints, power=money+noise). I don't think people who benefit from quad cores (video encoding, professional graphics manipulation, scientific computing, etc.) will complain about what you get for the price. The dual core 45nm chips were also impressive in power consumption and performance and attractively priced too, more appropriate for most classes of user.

I'm not a partisan (780G is great) but Intel made another big step forwards in 2008 and bit-tech chose to ignore that in favour of a 2007 part that is inferior in power consumption and performance.

[Tim S]

Regardless of whether Intel made another big step forwards from a technological standpoint, which we don't dispute - see last year's awards - the mainstream 45nm quad-cores were horribly unavailable in the UK and nobody at Intel would say why they were so hard to get hold of. And when you could get hold of them, they were over priced - that's what made the Q6600 so attractive in the UK.

The 45nm quad-cores were announced at CES and still had poor availability in the UK when IDF China was going on at the start of April. Later in April, we were told to "come back in August" if we wanted to buy Q9450 or Q9550 parts.

The Q9300 isn't as attractive as an overclocking chip because of its high FSB and low multiplier - the Q6600 hitting 3.0GHz was simply a case of increasing from 1,066MHz to 1,333MHz (something every board can do), whereas the Q9300 required a bit more effort and a board that could do at least 1,600MHz front side bus speeds. New boards designed with quad-core overclocking in mind can do that fine, but many cannot without a lot of tweaking.

When you're going beyond 3GHz, you need 480MHz FSB to hit 3.6GHz on a Q9300 (not easy with even the most astute P45 boards) compared to 400MHz on the Q6600. That makes quite a difference and 3.6GHz isn't unheard of on a Q6600 with decent air cooling.

[RobertBruce]

I agree its crazy to ignore the 45mm quads. Q9450 vs. Q6600?

[CSMR]

Point taken re UK issues. It's certainly annoying to live in the UK as a consumer.
Overclocking I don't think is a good argument. Why take a relatively high-power chip and overclock it? A 50% overclock gives >200% increase in power consumption assuming V proportional to f and a power=V^2*f relationship. This is just to show off and I can't imagine a serious use of taking an old chip and overclocking it. I suspect most overclockers take more time overclocking than they save in having a faster computer. And if you want to show off, pay $$$$$ and get the best.

[Jodiuh]

I love my Ikari, but the build quality between the 2 I've had IS terrible. The first one stopped being recognized due to some sort of cord issue and the second's feet scrape across the mousepad. The feeling's akin to a tiny dust rock underneath the mouse at all times. Also, both mice exhibit the problem of activating any or all of the top mouse buttons when the entire mouse is picked up and brought down firmly during a heated twitch MP shoot fest like COD4. This inevitably leads to a grenade firing off after you respawn. Also, the software does not work at all when trying to update the mouse with profiles, so you're stuck when trying to emulate a key for a game like Dead Space that doesn't recognize mouse 4/5.

Further, none of my questions were answered by Steelseries in my recent RMA. I love the mouse, but it has serious issues that need to be fixed if it's to keep up with my 4yr old MX518 that although will overheat if pushed hard at a LAN party, does not break down quite so easily.

[notatoad]

Originally Posted by CSMR

Why take a relatively high-power chip and overclock it?

because as enthusiasts, that is something that we like to do. these are the bit-tech hardware awards, not the best-computer-for-my-grandma awards, and to the readers of bit-tech in general overclockability is an important feature. power consumption be damned, i want my speed and i want the satisfaction of knowing my computer is running faster than the computer i paid for.

if i were to buy a new processor today, it would be a Q6600. but i won't, because my overclocked E4300 still serves my needs.