New RDRAM chipset "beats DDR by 50%" claim

[FishTankX]

And DX2 player, i'm 'Upgrading' because of the limitations of 256MB of RAM. I'm still perfectly happy with the performance of my PC800, and my Williamette 1.7... my performance standards aren't sky high. I haven't touched a single upgrade on my rig for more than a year. But 256MB of RAM isn't enough to play anymore.

The reason why i'm 'Jumping ship' to the SiS655 is in my current situation (You aren't going to believe this, but I live in China and this changes things a bit)

It's cheaper to sell my board at and RAM at half the price I bought them at, and buy a Gig of name brand DDR266+ SiS655 motherboard, than to keep and buy an extra 512MB of PC800.

I don't 'feel the need to upgrade' often at all. Heck, i'm still running a Williamette! I'd probably be perfectly happy with the performance of an i845D platform. My CPU doesn't eat all that much bandwidth, anyways.. I'm not happy if there's excessive swapping, though. That's what i'm trying to avoid.

My system specs are as follows.


1.7GHZ Willamette
P4T-E
256MB PC800 (128X2)
Radeon 8500 64MB
Baracuda ATAIV 40GB
Audigy

That was a pretty decent system a year ago. Haven't touched it until now.

I'll probably upgrade to a 3.06 hyperthreading northwood when it's cheap enough. (about 200$) and I think besides a videocard upgrade that's the last upgrade i'm gonna do to my system for a long long time. (Perhaps another year)

If you do buy 64bit PC1200, good for you. You have the money to spend on it and it will probably keep you happy for a long time.

Me, i've had my run with PC800. I've enjoyed it alot. Even if it wasn't worthit, I'm proud to have, at one point in time, bit the bullet and gone all out. I was told I was the first and last one to purchase an i850 board from that particular shop, that whole entire year. But I can't afford to keep up this game now.

Don't get me wrong. I like RAMBUS a whole lot. It's just that for the average guy, and the poor guy, it aint worthit.

I'll see how RAMBUS is doing after I get out of college in 4 years, and go all out for then. Till then, time to buckle down and start saving my penies. Hope you understand my situation better.

 

[NOX]

Originally posted by: DX2Player
LOL you buildt a PC800 RDRAM system 9 months ago, now thats not to bright, thats like making a single channel PC2100 system now and wondering why its not top of the line in 9 month, especially considering that PC1066 and PC800 cost the same you ripped yourself off and trying to blame RDRAM for it

AND FOR THE LOVE OF GOD ALMOST NOBODY WHO IS CONCERNED ABOUT $$$ USIS A GIG OF RAM, man your complaining about $ when your willing to buy a whole system for a few fps in 1-2 games, LOL if you care about $ so much the cheapest thing would be to just add more ram to the system you have.

Heeeeeyyyyy, cmon now I love my PC800 that I built over a year ago! I think it's still fast even though it's only powered by a 1.5GHz P4. I refuse to give it up in fact, if I ever replace it (which I know some day I'll have to), it will be replaced with another RDRAM system. Just like I replaced my i820 with this one.

I can't believe people still argue about this stuff, RDRAM vs DDR SDRAM what's with all the technology allegiance running around these threads lately?

You should never by because of your emotions to one company or brand, if you do let me say that you're missing out on some great technology that a lot of these other companies offer. I've tried almost all, VIA :|, AMD , Intel , and now Nvidia . nForce2 man, who would have thought I would be buying a chipset made by nvidia when the Riva TNT first came out. The onboard sound on this thing is amazing, features, wow what a chipset! I wish that they would make an nForce2 that supported both RDRAM and DDR, now that would be fun!

One thing I must agree with pertaining to RDRAM that has been expressed in this thread is RDRAM and its longevity, I mean, a year later and I still get a higher score with my RDRAM system then I do with my current every day system. That just shows you the extent to what RDRAM can do. It?s just an amazing peace of technology, and for that I respect it.

As far as the lawsuits, well let me just say I have yet to see any court rule against Rambus. Only time will tell what will happen to both RDRAM and DDRXX.

EDIT: OMG, my post looks like a xmas tree!

 

[Novgrod]

Personally I think it's of great importance to be a conscientious purchaser, so do my damndest not to support companies with awful business models based on extortion, monopoly, et al. I still don't buy intel in large part because of the first rdram fiasco, and because of their <ahem> suggestions that motherboard makers not produce for the very first Athlons. But that's just me, and I also do dumb stuff like giving money to the EFF.

Of course i encourage you to spend your money however you want.

 

[Ice9]

Not saying RDRAM is bad, Im just saying their success in the x86 space is at the mercy of Intel, and so far I haven't seen anything to indicate Intel will produce any more RDRAM chipsets for the i786/P7 family.
Maybe this will all change with i886/P8, and DDR will finally be pushed out of the market?

I agree with you 100%.

I think in the PC Market, Intel is key. If they don't accept, it will go nowhere in the PC desktop race.

The reason I feel it will happen, is tied to your next point:

As for financial factors, such as Micron going out of biz, I leave that to the fellas over at Intel, they didn't get where they are by being stupid, I trust them to do what it takes to ensure their CPU's aren't left without the proper memory technology, and I trust them far more than anyone on this BBS, Citibank employee or not(no offence intended Ice9, Im sure you understand what I mean).

[/quote]

The problem here is... It's not just micron. Intel went to Micron, Hynix, Infineon, samsung, etc, and OFFERED to refit their fabs to produce RDRAM for the P4.

All accepted with the exception of Infineon, Hynix and Micron - and then used this as an opportunity to keep control over the SDRAM market (and now, DDR). I've said this before, but what the hell, what's one more time Rambus' very business model is a threat to JEDEC memory manufacturers because it takes them out of the engineering loop. If it takes them out of the engineering loop, they have no control over how long they can keep their current fabrication methods in production. (hint: higher profits over longer use).

In a nutshell, they felt it necessary to try and edge Rambus out by dumping their products at ludicrous prices to keep it popular. This would slow Rambus adoption, and keep JEDEC manufacturers in control - and more importantly, KEEP RAMBUS OUT OF THE PC MARKET.

So, basically... Intel releases the P4, and their partners are PISSED because only one company is cranking it out - driving the price to $1200 per 128MB module. They SCREAM at Intel saying "GIVE US SDRAM and DDR, WE CAN'T KEEP HOLDING UP PRODUCTION WAITING ON RDRAM".

Of course, yields improved, manufacturing output finally met capacity, and all was well...

Until the memory companies agreed to keep RDRAM's prices artificially high. They figured they could keep this up until Intel could be persuaded to stick with DDR and move away from RDRAM solely on price. So far, they've won this "battle".

But there's a big problem here. Rambus didn't go away quietly And now, the very companies who tried to edge Rambus out of the market will have to sustain even WIDER losses in the form of more expensive royalties thanks to the infineon reversal.

See, here's the bigger, and greater problem. If all is hunky dory, patent law says you have to license your technology to the companies willing to pay out of good faith. However, patent law also says (in a nutshell) that you DON'T have to license them if they sue you and LOSE.

So here's the situation (and more importantly, what's at stake).

Infineon was the first to "take its chances" in court with Rambus. After a complete legal victory that was splattered all over the place (including their own investor websites), they have now witnessed something they never thought possible. A complete reversal by the CAFC. INFINEON THOUGHT THEY HAD THIS ONE IN THE BAG, until Rambus appealed. And won.

This clears the way for Rambus to continue with their lawsuit against infineon, only this time it's on Rambus' terms (using their definition of a bus, the whole markman ruling fiasco put back in their favor), and the next stop is supreme court.

For argument's sake, let's say you're infineon, and you're in this position. The only way you could win the first time around is to suppress the markman ruling that proves Rambus had defined the bus structure for SDRAM back in 1990 (2 years before JEDEC claim they did). Now, the appellate court completely reverses the ruling against you, and Rambus now has the power to PROVE that they are the original patent holder in court.

You now have a HUGE choice to make. You can:

1. Settle out of court and pay the 3% royalty going forward for DDR/SDRAM, and settle on that pesky $1B in back royalties and HOPE that Rambus in good faith will cut you a deal if you agree to push RDRAM with its lower 1.5% royalty rate (which you KNOW they will do)...

Or...

2. Continue on and take your chances in court, knowing full well that if you DO so, you risk paying TRIPLE damages, and running the risk that Rambus may never license you again for any of its products, basically throwing you straight into bankruptcy with no lawful right to make SDRAM/DDR.

I know which one will happen Infineon's shareholders are NOT going to let the top brass keep this court case looming over their heads now that they're sufficiently embarrased over the reversal.

And oh yeah, keep in mind that ALL other Rambus cases were put on hold by the courts pending the outcome of the Rambus vs. Infineon appeal Now that it's over, the precedent has been set. Most people don't understand just HOW BIG of a win this was for Rambus.

 

[Sunner]

It will be quite interesting to see what will happen in the next 1-2 years.

If Intel actually had RDRAM anywhere on their roadmaps things would be very different in my mind.
What makes me so doubtful, no matter what happens with the whole Dramurai vs Rambus ordeal, is that Intel still only has faster DDR on their roadmaps, such as the up and coming DC-DDR400 chipsets.

Also, one wonders what AMD will do if the Dramurai utterly fail rather than come to some sort of middle ground to work things out.

Oh well, Im neither an analyst nor a top level exec at Intel, merely a poor sysadmin and techno geek, so I guess there's no point in worrying about what I can't possibly do anything about anyway

 

[Ice9]

Now, i'm not being sarcastic here. I'm as worried about hynix (Which produces some of the best graphics card memory on the planet) and Micron as you are. But why don't they all raise prices together? It's really their fault that they got into this mess, and I don't think us consumers are gonna get them out of it. What should we do? Ask them to raise prices?

I suppose that all depends on how much you actually care about the company

Personally I think it's not nearly as bad as theh time where you could get 512MB crucial DDR2100 for 30$. Micron, in the financial reports you gave me, is posting losses yes. But minor ones, recently. To the tune of 30 million.

Ahem. Read it again: Their operating losses are $316M Q4 2002, $587M Q3 2002. That is from the most recent 2 quarters of operation. I don't know where you come from, but nearly $1B is a LOT of money to lose in a scant SIX MONTHS.

One thing I don't understand though. If the i850E is priced around the same as the SiS655 and was priced way below the granite bay, why couldn't samsung just drop the prices a little, so that they could be more competitive with DDR?

2 reasons.

1. Competition from only one other company (Elpida) means THEY DON'T HAVE TO LOWER PRICES!
2. They NEED these profits to subsidize their losses in DDR.

Then it wouldn't be relegated to all but the ultra high end. PC1066 that is. It certainly wouldn't have adversley affected it's profitability. Is it possible that they were already producing at maximum capacity, so that they couldn't fulfill any more orders?

Like I said... RDRAM is ARTIFICIALLY expensive because DDR is selling at a huge loss. I can't stress this enough. While it's APPROPRIATELY PRICED for what it is, the reason it isn't CHEAPER is because there is VERY LITTLE COMPETITION. Believe me, if you had 13 dram manufacturers manufacturing RDRAM, It would DEFINITELY be cheaper.

And as for my statment on RAMBUS being a very litigious company, I was under the impression that they got their SDRAM and DDR patents shadily anyways, and not just that. It also sees that they sued alot of people in a very short amount of time. But i'm uneducated on the details on this issue. Care to explain exactly how RAMBUS got the patents for SDRAM and DDR? I still don't understand that too well. Some people say they stole it from a JEDEC confrence. Others say they got it legit. What's your take on it?

My take on it:

Rambus has done nothing but develop memory technologies since their inception. In 1990, Rambus applied for a patent for a bus type that very closely resembles what SDRAM became in 1992+, as a result of the JEDEC conferences.

Now, even the CEO of Rambus admits that the patent was amended to include some of the things that were discussed at JEDEC. However, the bottom line is: so did everyone else at JEDEC. IBM did it. Intel did it (and even went on to sue VIA over some of it). Rambus did it. Oh, and if you read the judge opinions from the CAFC on the rambus case, INFINEON did it too.

Now, 2 of the 3 judges that made up the majority vote of the CAFC panel illustrated that JEDEC's own rules didn't cover the committee's policy on patents adequately. They also noted that all the other companies in JEDEC were doing exactly the same thing.

"Shady"? Maybe. Business always is.

"Illegal"? Nope. Not by a long shot.

"Fraudulent"? Not if EVERYONE ELSE was doing the same thing... If it were, then everyone in JEDEC is guilty of fraud.

Now, the funny thing is - the "amended" portions of rambus' originally filed 1990 patent ISN'T what rambus needs to prove that they own SDRAM. The original markman ruling is based on the ORIGINAL FILING of Rambus' 1990 patent. NOT the amended portions in 1992. This is why Infineon NEEDED to supress the markman ruling before trial, which they did during the first trial. There was NO WAY Infineon could win with that pesky 1990 patent being out there that clearly shows Rambus thought up SDRAM first. Of course, the CAFC panel KNEW about the markman ruling being reverse, and saw it as a pretty shady tactic as well. Why suppress valid evidence in such an important trial?

In the end, Rambus won, and they're now free to collect HIGHER royalties on LOWER technology patents of which they are the rightful owners. So, in the end if DDR wins, it doesn't matter to Rambus. They collect bigger royalites regardless.

Of course, Rambus wants to push RDRAM which has HALF the royalty costs and FAR better scalability. Oh, the horror

 

[Wurrmm]

Me...I could care less about what the buisness people are doing. All I want is the best bang for my buck when I upgrade in the spring (May/June). Where I go with my upgrade after that will be based on the same criteria. What ever is better at the time RDRAM or DDRII, Intel or AMD, Nvidia or ATI.

 

[dexvx]

Originally posted by: kuk
I'm surprised no one has said this ... but ...

GUTB??? :Q

Just because he's a supporter of Rambus doesnt make him GUTB. From what I remember GUTB really loved bashing AMD, and thats what got him banned.

 

[Ice9]

I would hope that being FACTUAL doesn't get you banned here

 

[fkloster]

I applaud u Ice9 and DX2player......where were u guys 2 years ago?

 

[codehack2]

This thread has confirmed my suspicions that JEDEC is nothing more than a front for Colonel Sanders and his concealed attempts to gain strong arm control of the PC industry. After having conquered the fast food industry with his so called "Secret Recipe", the Colonel is now enacting plans to slowly infiltrate, and take over the PC industry a market at a time....

To quote an old/wise Scottish man... "Oh, I hated the Colonel, with his wee beady eyes and that smug look on his face - ?Oh, you're gonna buy my chicken! WHY you ask! Because he puts an addictive chemical in his chicken that makes you crave it fortnightly, smartass!"

God help us all when his plan fully unfolds.

CH2

 

[Ice9]

I applaud u Ice9 and DX2player......where were u guys 2 years ago?

Hey, i've been a Rambus supporter from the get-go.

I knew they were on to something back in the PC600/711/800 days, and seeing them continually engineer top-rate platforms consistently tells me one thing: All roads lead to Rambus.

It's easy to bash a product you really know nothing about. It's just not my *personal* style.

Like i've always said. It's perfectly OK to dislike a product, as long as you're RIGHT about it. Most people who hate Rambus or feel their business practices are unethical don't know what the company is all about.

There's a ton of misconceptions about this company, and 99% of them are simple lies and fabrications. Even this Evan Lieb guy who reviews for anandtech is guilty of believing these misconceptions. And so far, he hasn't come forward to discuss this issue

I get the feeling he won't It's hard to argue your point when you have Micron's Crucial ads splattered all over your website.

 

[Ice9]

Cmdrdread: Quick question for you:

You mentioned that SIS' 655 dual DDR chipset supports Hyperthreading...

But according to the below article they haven't yet licensed it, and don't plan on putting it into a chipset until they sample the new 65X's in May/June:

http://www.theinquirer.net/?article=7981

Could you clear the air on this one for me? Or were you just BSing without fact? Or worse, is this chipset not even going to be AVAILABLE for months?

And in response to your "Now you've gone too far, who are you to question, blah blah blah" rant, I actually HAVE done some benchmarks of my own machine. Visit THIS PAGE to check them out.

 

[codehack2]

Originally posted by: Ice9
Cmdrdread: Quick question for you:

You mentioned that SIS' 655 dual DDR chipset supports Hyperthreading...

But according to the below article they haven't yet licensed it, and don't plan on putting it into a chipset until they sample the new 65X's in May/June:

http://www.theinquirer.net/?article=7981

Could you clear the air on this one for me? Or were you just BSing without fact? Or worse, is this chipset not even going to be AVAILABLE for months?

I love it... your qouting the inquirer as a source and accuse someone else of BS.

I assume that Gigabyte listing thier 655 boards and bios revisions with HT support is proof enough?

Gigabyte SQ800 & SQ800Ultra HT Support

CH2

 

[Ice9]

Originally posted by: codehack2

Originally posted by: Ice9

I love it... your qouting the inquirer as a source and accuse someone else of BS.

I assume that Gigabyte listing thier 655 boards and bios revisions with HT support is proof enough?

Gigabyte SQ800 & SQ800Ultra HT Support

CH2

Eh, if I believed the inquirer's word as gospel, I wouldn't have even asked.

Anyway, just the info I was wondering about. But if you read the footnote for those boards, it says in plain engrish:

*At this stage, due to the limitation of SiS chipset (655(A0), 648(A2), 651(A1), and 645DX(A0)) , there will be some circumstance when Intel Pentium 4 Processor with Hyper-Threading Technology is applied and enabled. Please see the manual for more detail.

Care to translate? My engrish skills are poor. But to me, it says "It might work, it might not". Considering there's only ONE CPU out there in the P4 family that supports HT, just WHAT does the above footnote mean?

 

[codehack2]

Originally posted by: Ice9

Originally posted by: codehack2

Originally posted by: Ice9

I love it... your qouting the inquirer as a source and accuse someone else of BS.

I assume that Gigabyte listing thier 655 boards and bios revisions with HT support is proof enough?

Gigabyte SQ800 & SQ800Ultra HT Support

CH2

Eh, if I believed the inquirer's word as gospel, I wouldn't have even asked.

Anyway, just the info I was wondering about. But if you read the footnote for those boards, it says in plain engrish:

*At this stage, due to the limitation of SiS chipset (655(A0), 648(A2), 651(A1), and 645DX(A0)) , there will be some circumstance when Intel Pentium 4 Processor with Hyper-Threading Technology is applied and enabled. Please see the manual for more detail.

Care to translate? My engrish skills are poor. But to me, it says "It might work, it might not". Considering there's only ONE CPU out there in the P4 family that supports HT, just WHAT does the above footnote mean?

Translation for the english impaired... It means the B0 stepping of the 655 officially supports HT and they have released a bios that will in most cases allow A0 stepping boards to support HT.

From The AT review of the sinxp1394

2-12-2003 UDPATE: Gigabyte sends word that, even though the vast majority of SINXP1394 motherboards will be B0 stepping, there will be a small amount of A0 stepping SINXP1394 motherboards sold too. Gigabyte says they have made a particular modification to the BIOS of A0 stepping motherboards that allows full support for Hyper Threading (all B0 SINXP1394 boards support Hyper Threading of course). We have also received information that, while the SINXP1394 is capable of 800MHz FSB support through overclocking, Gigabyte cannot guarantee that the upcoming 800MHz FSB Pentium 4 processors will operate correctly in Gigabyte SINXP1394 motherboards until they find out exactly what changes Intel makes to mass production 800MHz FSB Pentium 4 silicon.

We will keep you updated on 800MHz FSB support as we continue to receive information throughout the month.

CH2

 

[Ice9]

Thanks.

Now if we could only get a proper benchmark of that chipset with a 3.06

 

[codehack2]

Originally posted by: Ice9
Thanks.

Now if we could only get a proper benchmark of that chipset with a 3.06

Ask ye shall recieve...

VR-Zone review of the AX45-4D (sis655)...Here it is again at Sudhian

CH2

 

[Cerb]

OK, the topic drifted a good bit, so I have just been reading, but...

AND FOR THE LOVE OF GOD ALMOST NOBODY WHO IS CONCERNED ABOUT $$$ USIS A GIG OF RAM

You are quite wrong. I know a few people who wnat 1GB of RAM and would prefer to get it by saving money and using PC133. While I'd think if you're going to do that you could at least go with a PC2700 (coupled w/ KT333, NF2, GB, 648, 655, 845PE/GE) ...I'm not them. Anyone worried about top of the line isn't worried about cost. Likewise, anyone worried about cost isn't worried about top of the line, and has specfic needs that take precedence over others. If that weren't the case, cheap old things like the K7S5A wouldn't still be selling so well.

 

[DX2Player]

Originally posted by: fkloster
I applaud u Ice9 and DX2player......where were u guys 2 years ago?

Honestly 2 years ago I didnt know jack about the inside of the computer infact 2 weeks before 7 months ago I didnt know jack. But what drove me? Well basically my off to college computer my parents got me was so inferior I found myself searching for a new one. It was the internet special with no AGP slot and SDRAM. Over the summer my friend built a computer of his own, I never even imagined making a computer myself till that point but figured it counldn't be that hard. So spent two weeks reading 24/7 about every computer part on every review site I could find to build my high performance $1000 machine. And well frankly I became infactuated with it and never stopped learning since. I have found the best deals and was looking for a machine to carry me through the rest of my college years. I might say I wanna get this and that but honestly I have no money and will probably do nothing but make small upgrades to this computer. Four years at GT for an out of state student is not cheap.

Being that I have so recently been on both sides of the fence (being average user to being an informed user) I know how the average person thinks. The average user doent know what type of ram is in his computer, the average user wouldnt think of building a computer, the average user would never overclock, the average user would never buy a gig of ram. So what does the average user know, the average user knows a bit about video cards, the average user knows that more MHZ = better, the average user knows that RAM means random acces memory.

 

[Ice9]

Ask ye shall recieve...

VR-Zone review of the AX45-4D (sis655)...Here it is again at Sudhian

Thanks for the links!

On VR-Zone, they list 2 test rigs... a 2.4ghz and a 3ghz with Hyperthreading. They appear to be running only one set of benchmarks. Which CPU are they using? Any idea? Their test setup doesn't explicitly say which CPU they're using for their benchmarking tests.

I can pretty much assume they're all on the 3.06HT, since they seem to be right around [L=my own benchmarks]http://www.weckstrom.com[/].

Weird, though.. on BOTH their PCMARK 2002 tests, their memory tests are significantly slower than mine (across all platforms). Maybe having 1GB makes a difference? *shrug*

Either way, according to those benchmarks, SIS dual channel DDR has caught up to 850E. And it's, err, well, about time I guess

Now, bring on Redwood, PCI Express and Yellowstone, for god's sake.

 

[First]

Thanks to codehack2 for bringing this thread to my attention. And thanks again codehack2 for linking to articles that clearly show SiS 655 in dual DDR400 mode is faster (but only slightly) than 850E in dual channel PC1066 mode, it saved me some trouble.

Just as I said many weeks ago, our own in-house tests proved that SiS 655 and 850E boards running a CPU just over 2.8GHz made no difference in performance versus the 2.26GHz CPU used in our three SiS 655 reviews. In fact, just this past weekend I conducted those same benchmarks with SiS 655 and 850E boards running at 3.2GHz with HT enabled, and found the SiS 655 to be even faster at 3.2GHz (dual DDR400 mode) than it was at 2.26GHz (dual DDR400 mode) when compared to the 850E in dual channel PC1066 mode.

So tell me again why you believe that a higher CPU speed with HT enabled would make 850E faster than SiS 655 Ice9? Is it your, apparently untested, theory that the ?memory subsystem will be stressed at higher CPU speeds? and therefore 850E is faster than SiS 655? (have you even tested an SiS 655 board yourself?) Or is it like a similar theory, which states that a 533MHz FSB P4 doesn?t gain any performance by running in dual channel DDR333/400 mode because anything higher than dual channel DDR266 exceeds the peak bandwidth a 533MHz FSB processor is able to handle, therefore lowering performance because of latency penalty? Our results (and other web site?s results) prove otherwise.

I have no problem with criticism Ice9, but at least do your own testing if you?re going to claim that 850E will be faster than SiS 655 at 3.06GHz (HT enabled) and then tell me I'm wrong for using a 2.26GHz CPU that doesn't "prove" your untested theory.

*Edited for last part*

 

[First]

Originally posted by: Ice9

Ask ye shall recieve...

VR-Zone review of the AX45-4D (sis655)...Here it is again at Sudhian

Now, bring on Redwood, PCI Express and Yellowstone, for god's sake.

Agreed. But don't forget FlexPhase.

 

[FishTankX]

Ice9, SiS has had a hyperthreading liscensce for two months.

SiS announced its hyperthreading liscensce around christmas!

 

[Ice9]

Fishtank, thanks

I'm glad they have that license. Can't wait to see how well the 658 does. Though in all honestly, I probably won't buy it

I'm just not much of a believer in 3rd party chipsets... I simply hope it performs well enough for Intel to go back to Rambus. I have no doubt it will outperform anything else available, but stability and compatibility mean everything to me. It'll have to hit a few of my friends' desktops first