Slashdot Mirror
AMD Desktops Outsell Intel
prostoalex writes "For the week ending August 21st AMD managed to capture 54% market share among new desktops sold. Intel's share during the week was 45%. While Intel leads the U.S. CPU market with 82.7% market share, folks from AMD are proud to announce this is the second week this year - they also outsold Intel on the desktop market one time in April 2004."
I'm not sure I buy this. Maybe for enthusiast and home gaming PCs, but if you include business desktops I'd venture to say that Intel still carries somewhere around 75%. Go look at the business-oriented desktop lines from HP, IBM and Dell and you'll see very few AMDs in there.
Duboise continues: "promotions continue to be the driving force behind retail PC sales and AMD's successes. In fact, $699 notebook promotions have been the driving force behind three incidents this year when notebook sales were able to overcome desktop sales. As long as Intel continues to place more emphasis on the more lucrative and successful notebook market, it leaves the door open for AMD's desktop wins."
I wonder if they believe that they can eventually drive notebook sales upward to the point that they outsell Intel more often than a handful of times a year?
Cheers,
Erick
http://www.busyweather.com/
First ATI outsold Nvidia on desktops, and now this! Good to see theres not a monopoly on core hardware components! now if only software were the same way, :\
Nobody could ever need more than 7.314 MHz of speed.
Intel is planning to release a 10.000 MHZ cpu and kick AMD poor lame ass.
Watch it, in several European countries (and a bunch of non-European ones) that truly does mean 10,000 MHz :)
We've come a long way from the "AMD is Dead" and "Intel Rules" days.
Intel let its marketing people get caught napping. Intel pushed the Itanium and said it will never make a 64-bit chip that is x86 compatible.
AMD came out with the 64 bit chip that was compatible with the x86, and it got rave reviews. And, it gets sales!
Now, AMD outsells Intel again. Did you see that -- the article said "again."
Not bad for a company that was being written off a couple years ago.
I was an ardent fan of intel until the Athlon 64 came out. My brothers new PC has an Athlon 64 along with other goodies (1gig ram, dual layer dvd writer...) for a very reasonable $1,000 USD.
There is no way I could have done that with an intel chip and motherboard and still get the same performance.
What could possibly go wrong?
If you can't beat 'em, change games.
The VIA KT600 chipset is definitely slower than the exceptional NForce 2 chipset.
It is also cheaper.
Anything Intel blows goats compared to a decent Nforce 2 board.
Are you listening?
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
That was total bullshit, not insightful at all. Anybody doing serious audio work on an X86 PC, even at the low end of the market, is using a breakout box for I/O and conversion.
Whether quality or no, the major thing AMD has going for it is price. For those of us non-gamers, who are fine with a pretty darn good machine, even sacrificing a little quality for a major difference in cost is worth it.
Snazzier than a Three-Piece Suit: http://kf.rainydaycommunications.net/
The thing about monopolies is that, as in a business sense, it's the equivilent of launching the space shuttle at the end of Civilisation - you win, you achieved the ultimate goal, and every company wants to be in that place.
Therefore, behind a monopoly, there's always another monopoly watching and waiting in the shadows, looking to take over from the popular and dominant market force. For example:
Microsoft (Apple)
Intel (AMD)
Nvidia (ATI)
iTunes (Napster)
IBM (Sun)
Gnome (KDE)
They'll tell you that they wouldn't ever want to be a monopoly, but it's poppycock - the whole principles of business are based on monopoly.
Sunday you're Thinking Different, Monday you're a huge tool, paying too much and waiting to think like everyone else.
AMD is going HT one better by putting two Athlon 64 cores on one die next year. Much better performance bump than HT provides in a single processor core.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
Uh, a theoretical 4-fold speed increase isn't anything to laugh at. Ask AMD, Intel, and IBM if they'd like to have a new idea in microprocessor design that had a theoretical 4-fold speed increase, and I'm sure they'd all say: "Yes."
I feel bad for Intel and HP though. Sunk quite a bit into making sure Linux could run on the Itanium from the very start, getting little for it. Really, the Itanium and Opteron are like apples and oranges.
One major problem is gcc. GCC just can't handle EPIC stuff yet. The compiler from Microsoft, Intel and HP are quite a bit much better than gcc. But gcc is the defacto complier in Linuxland, like it or not. (Even though Intel's x86 compiler for linux can do better too).
One of the major complaints I have about my XP2500+ is that the thing runs hot, like really hot. We hit a heatwave locally and temperatures were up to about 40 celcius at peak. My CPU actually hit 95 celcius (for those that use Fahrenheit, 100 celcius is boiling temperature).
I have a bigass thermaltake fan in there now, which I can turn down when the weather is cooler. The computer is still rather noisy.
My point to all this is not AMD bashing however. Apparently the 64-bit CPUs do much better for heat dissipation. The CPU die is much larger (the actual die is small on an 32-bit Athlon), so heat dissipates much more nicely into the heatsink due to the increased surface contact area. When I do upgrade, I'll be going AMD64... more power (in 'nix anyhow) and cooler running than my current CPU.
Anyone who cares about using their computer for professional audio applications, for one.
Funny, we seem to have rather a lot of pro audio users on Macs, last I checked.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
In my experience, AMD Athlon CPUs tend to perform better at video editing which is why I chose an FX-53 over a P4EE - it has much better memory bandwidth due to the memory controller being on the CPU instead of the northbridge and the FPU performance is usually superior.
Of course, there's always the 64-bit thing to take into consideration (and the no-execute extensions in Windows XP SP2).
Intel themselves admit that the only way they could make 64-bit desktop chips was by copying the AMD 64-bit extensions.
Professional Audio applications aren't running with onboard AC97. You'll have added a high-end card, or two, to your system. The only way the N/S bridge chips could add crackles would be if they weren't exchanging data properly, in which case nothing would be running correctly on your computer.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
This is a 100% bonafide GOOD THING. Have you seen what these guys have done to each others margins? Have you seen how fast processor speeds have become these last 4 years? This is competition at its absolute finest.
Cheers to AMD for not giving up and dying. And cheers to that chairman of theirs who looks like he oughta be out selling chicken.
Well, it could have been anything; the video cards, the motherboards, the RAM, the power supplies, inadequate cooling or ventilation, who knows. Who put together the computers?
I am very doubtful that the actual CPU was the cause of any instabilities.
Competition means better products at lower prices
Almost always, competition means lower prices. It does not always mean better products, unless you mean better product for the dollar. If a product were a third less reliable for half the cost, then I suppose that is a win, if you don't consider environmental impact of each respective product.
As for better product, it doesn't always seem to be the case, at least as often. Sometimes there is some corner cutting on the part of all competitors, note the quality or availability of support lines, flimsier material or the missed testing step in a rush to get the product out to beat the next company. Many consumer electronics seem less reliable these days.
Ummm...
Last time I checked I used AMD for pro audio applications. In fact, I've won more awards than many other audio producers I know (not that it has anything to do with AMD... or doesn't or whatever). My 700 mhz PC worked fine. I still use a 1.2 GHZ AMD PC as well too. I've done some corporate video work on both PCs as well.
Unless Intel gets better and cheaper, I'll be sticking with AMD.
Honestly, I just built a desktop and didn't even consider Intel processors. It seemed that all the boards I wanted were AMD.
Coincidence?
I wouldn't say that Intel has everything wrapped up. comparing a Xeon to a 3800+ is hardly fair as you are comparing a server processor to a desktop one. Now if you compared it to say, an Opteron (a much more fair comparison), well then you'd see AMD still wins or pulls up even.
What's more, the more processors, the better. Hypertransport gives each processor it's own bus.
That said, comparing an FX-53 to a 3.8 GHz Intel would also be a more fair comparison. And while it's true that the Intel wins it's share of benchmarks, keep in mind: You are comparing a 3.8 GHz Intel chip to a lousy 2.6 GHz processor (the FX-53). Theoretically, the Intel should totally kick it's ass - but it doesn't. That's some good chip design there my friend!
I just got a 3800+ last week. All I can say is: WOW!
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
With that many digits after the dot, what number system are you using? That could be either 10 GHz or 10 Mhz, depending on what kind of audience is viewing...
I wonder if they believe that they can eventually drive notebook sales upward to the point that they outsell Intel more often than a handful of times a year?
The article says that AMD's desktop successes are partially a result of Intel's tendency to emphasize notebooks. If "they" (Intel, I hope you mean) drive notebook sales upward, and assuming that damages desktop sales, Intel's sales would increase because of their notebook dominance and AMD's would decrease because of their desktop interests. Overall the desktop market would shrink (or grow less), while AMD's share of it might grow marginally as a result of the notebook market distracting Intel from pushing its desktop CPUs as aggressively. We might then have more "AMD Desktops Outsell Intel" stories, but it would definitely not be good news for AMD.
What exactly does EPIC provide, apart from a lot of work for compiler writers and a (theoretical) maximum 4-fold speed increase with current designs?
"All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us?"
Friends don't let friends misuse the subjunctive.
" I'll never buy another VIA-based mobo due to all the problems I've had with audio."
Soooooo, , don't buy another Via based mobo then. It's quite simple really.
What has this to do with AMD CPU's again?
"You can't fight in here, this is the war room!"
I'd like them to make a dual core Pentium M, add the latest SSE stuff (SSE3?), an on-die, dual channel memory controller, a HyperTransport bus, and sell it for the desktop crowd.
Is that too much to ask? *sigh*
I have an N2U400A from ECS. It is a cheap ass mobo, but when I get it installed and running I was surprised at how sweet the NForce 2 Ultra chipset is. Fast, and I do mean FAST and stable. I put some stress on my system. I'll run the distributed.net client in the background, while I have a VPN connection open to the database server, I'll be importing records into that database, A VNC session running, and all while while playing Counter Strike in the foreground. System stays smooth and responsive.
Via needs to change something, because the NForce chipset is kicking ass all over their offerings for only a few dollars more.
I was kind of wary when I heard that NVidia was releasing a mobo chipset, I thought that since they were a "video card" company that they wouldn't be able to make a good chipset. I am so glad I was wrong. Now, I'd like to see what ATI can do.
LK
"Hi. This is my friend, Jack Shit, and you don't know him." - Lord Kano
While AMD's x86 stuff is better than Intel's x86, its a bit sad that Itanium has lagged so far behind Opteron. Itanium's architecture is vastly superior to Opteron's, as it marks a break from the 20-year accumulation of old designs and legacy crap. It would be nice to see people embracing a new architecture for once.
Or course I can't claim superiority, having purchased Opterons myself. I guess software availability will always win out over good chip designs. Just ask those poor Alpha designers over at Compaq/HP.
Don't you hate meta-sigs?
where are all those intel favourable benchmarks?*
lots of amd favourable ones
in my personal experience, Intel's always have a small lag that is quite noticeable. Although this is comming from the same person who can tell a 85hz refresh rate from a 75 so its probably not something most people have to worry about.
and THEN there is the huge price difference
*(i wouldn't personally count office benchmarks like word but i know intel has a weird history of doing well in those)
I'll just use my special getting high powers one more time...
Funny, I've never had the problems you've described short of bad configs or bad boards. Shall we cite chapter and verse of Intel's mistakes lately? WE don't even have to talk about direct mistakes. The 810/815 series isn't exactly what I would call 'stable'. All things being equal there are ALWAYS cheaper alternatives that aren't quite as stable from Intel, VIA, SiS, and Nvidia. That's why you plan to get the better ones when stability is a must.
Last week I purchased a NEO2 board (NFORCE 3) and a 3800+ Athlon 64. It TOTALLY rules. Know what else? I do video and 3D rendering - a LOT of it. I have it OC'd about 5% so far too. No lockups. No 'crackles' from the Audigy. Everything is cool and I can't believe how frigging CHEAP it was!
"...Well, there's egg and bacon; egg sausage and bacon; egg and spam; egg bacon and spam; egg bacon sausage and spam..."
Uh, this is a VERY well known problem with Via 686B Southbridges (among others) and Creative labs soundcards. Neither company knows how to follow the PCI 2.0 or 2.1 spec and so burst data transfers done by the sound card are corrupted. Some Firewire cards also have problems transfering to the iPod for the same reason. The problems were enough to put me off VIA permenantly. I now use SiS chipsets for my AMD systems and have had no problems (I don't need any of the expensive integrated stuff from a NForce board).
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
The purpose of HT is to make up for Intel's crappy super-long pipeline (something like 32 stages!? Someone correct me if I'm wrong). Whenever it does a jump, many instructions are wasted. AMD's pipeline on Barton was something like 12 stages, so there's much less wastage going on. All HT does is allow the wasted cycles to be used for another thread. Since AMD's processors don't waste so many instructions, HT wouldn't really help that much.
Don't you hate meta-sigs?
(hint: they're actually innovating)
honnold.org - sometimes-rock band, all the time awesome forum
I posted in a reply that doesn't appear to be getting modded up, so:
The figures for Intel's total share are worldwide, not US. (I should know, my company is the source cited in the link.) Meanwhile the AMD weekly share data (from another company) is for US Retail system sales. So the two data points really aren't comparable on any basis.
I know the figures I cite are exclusive to x86 CPUs. Someone mentioned PowerPC in this thread, and Apple provides sales figures as part of their financials -- based on Q2 data, PowerPCs in Apples comprise about 1.8% of the market if you included them in the calculations.
As I understand it (which may be fulla holes) the N/S bridge chipset matters a lot. Professional audio apps are notorious for having problems on the PC platform and the problem is that the PC platform was not designed with realtime (or even psuedo realtime) constraints in mind.
Even if you have a pro audio card that does a/d conversion, the data still has to get from the card to HDD fast enough. The system is probably way more than fast enough on average, but you get pops anyway if some other process keeps the cpu busy long enough for a buffer somewhere to fill up.
The chipset is key because audio is much more i/o intensive than compute intensive. So, the bottlenecks are definitely on the i/o bus (or maybe memory bus? I dunno.). I would guess that any pro audio app will have code that's been hand tuned to work with the patterns of latency typical in intel hardware.
But still, cheers to AMD for kicking some flabby, complacent, celeron-crippling, market-segmenting, mhz-is-everything intel ass.
-chris
-cbare
Very true. Also, Intels seem to be more vulnerable to the dreaded P4 denormalisation behaviour (whereby many older plugins cause massive cpu spikes when not being used), which is a Very Bad Thing for audio work.
in the last 5 min?
Whenever it does a jump, many instructions are wasted
No. Branches generally do not cause a pipeline flush. (This is why branch prediction is a hot topic.)
HT does not exist to operate only in pipeline stalls. HT exists because analysis demonstrates that most x86 programs do not exhibit enough parallelism to fully utilitize all of the multiple execution units in a modern Pentium. You've got a lot of silicon devoted to peak performance that isn't used all the time, because you don't happen to have (for example) a bunch of full-width add instructions going on at the same time. HT allows a second thread to use those chip resources.
HT is cheaper than building two processor cores, as lots of the instruction fetch and decode logic is shared. Putting two complete cores on the same die does not increase the efficiency of utilization of the resources in either core. Dual core is much more of a brute force solution to the problem (a complaint AMD fans usually lodge against Intel). In this case, execution units in both cores will often be idle, as neither thread alone happens to need the full capability of a single core.
Since you've spent more silicon on the problem, dual core can have performance advantages -- specifically whereever you actually need that duplicate logic that would be shared with a HT design. Often, however, that extra fetch/decode logic is going to waste as well.
HT is an elegant optimization for a modern superscalar processor. It is not, however, the same thing as a dual processor, nor does it solve exactly the same problem.
Coincidence? Naah...
HT (Simultaneous Multi-Threading) won't benefit AMD's chips as much as it does Intel's because of the way they are constructed.
/.)
Due to the P4's incredibly long pipeline (30-odd stages?) and very high clockspeed, if the branch prediction goes wrong, the chip will stall
HyperThreading is a clever hack that runs two simultaneous threads on the same die. In this way, if one thread stalls, the other can execute in it's place while the other thread waits for the pipeline to redo itself, hence being a very clever way of making up for the design "faults". AMD's typically run at a lower clockspeed, and have a much shorter pipeline, so even when their piplines stall, the chip does not waste as many cycles - in short, they;re not really designed to take advantage of SMT. Hence AMD not having SMT support is a bit of a non-issue.
(Disclaimer: I'm not much of a buff on chip architecture, this is just stuff I've picked up from reading
Moderation Total: -1 Troll, +3 Goat
*I don't like KDE all that much either, but then again I'm also not a big fan of GNOME
"[Regarding the 'cloud,'] ownership was what made America different than Russia." -- Woz
- Intel still sell more CPUs than AMD
- My friends' AMD blew up so I'm never using them again
- The Celeron is a 'fine' processor
- People keep mentioning that VIA make CPUs (chuckles
;-)) - No-one gets to the point!
Being that competition is a GOOD thing - why Intel is allowed to bribe Dell into supporting its massive monopoly is mind boggling. Another thing that I don't understand is this - if Dell were to swap CPU lines overnight, from Intel to AMD, no-one would care because its the Dell brand that matters not the CPU maker - I suppose its just about volume - you can bet if AMD could guaratee supplies to Dell they'd do it because it would be good for profits, whatever deal they are getting from Intel.I remember building a Z80 single board computer when pursuing my electronics degree.
The Z80 CPU was so stable that we could actually hook a potentiometer up to the timing circuit and scroll the system clock speed up or down and it just went on about its business of running happily. As I recall, it maxed out about 1 megahertz, but you could reduce it way down without trouble.
this doesn't include systems that people build for themselves, which more often than not are built around AMD processors.
Uh oh.....time to start hating AMD.
Qxe4
Yeah, HT works better on deep pipelines, but it's still a win for shallower ones. I couldn't find any more recent figures on AMD's pipeline depth, but duron was using a 20-stage pipeline.
That's not shallow.
I saw a graph based on HT simulations, it was CPU utilization versus degree of hyperthreading. Utilization didn't start to level off until you have something like 6-8 execution contexts on the chip (that's assuming you have work for all of them, of course.)
That's probably geared towards an Intel-style pipeline, but surely AMD would benefit from adopting the technology. Though since they're already going with dual-core, I'd rather see them apply it towards speculation to absorb cache misses than expose those contexts. That way single-threaded apps could get some love.
There's no way in hell SMT gives you an 80% speedup; on the P4's it gives 10-15% at best. Thankfully, it comes at little-to-no-cost in terms of silicon, IIRC. SMT is nowhere near actually having two physical processors, it just utilises the power of the P4 more efficiently by just filling in the gaps in the execution stages. Some apps actually show a performance decrease under SMT.
Dual cores won't give you 200% either, even with the Opteron arch it'll still be 190% at the highest.
Moderation Total: -1 Troll, +3 Goat
So it sounds like AMD is going without HT.
Damn.
They may not have a choice (patents).
HP's computer line is a joke everywhere. Well, everywhere outside Elbonia.
"We have to go forth and crush every world view that doesn't believe in tolerance and free speech." - David Brin
Smaller engine=less area=less heat.
That rule would seem to be inverted for CPUS.
Actually, it's not.
Power consumption in CPUs is dictated by clock speed, die size, and feature size (90nm, 130nm, 180nm, etc.).
If you were to take the old Pentium II design, and re-engineer it for the modern 90nm process (or better yet, the upcoming 65nm), you'd be able to shrink it down to a smaller die size. This would yield both a smaller physical chip (which would be cheaper to produce because it's using less silicon), and lower power consumption, assuming you ran it at the same 300 MHz or so that the old P2s ran at.
The problem is that chip companies and consumers don't care about lower power consumption; they want faster performance, or more precisely, they want bigger numbers so they can brag to their friends and feel like they're doing better than the Joneses. So while going to smaller feature sizes helps reduce power consumption, going to a higher clock speed more than makes up for it, so the actual power consumption is continually rising.
Even worse, with transistors becoming ever smaller, the heat they produce is being concentrated into smaller regions, which causes localized heat problems on the chip, necessitating more engineering solutions to keep those areas from overheating. If you look at a thermal map of a CPU in operation, you'll see that a very small part of the CPU is generating the majority of the heat--the ALU and execution units, which are constantly utilized, produce most of the heat, while the SRAM cache produces very little even though it probably accounts for a majority of the die real estate.
For plain GHz monsters, AMD simply is the better price/performance deal. Now lets look at the situation from a different angle. Intel has a kick ass processor in their line. Yes the Pentium-M faster than most except the high end P4s only sold in servers and laptop computers. Outside of the US there is a huge market for machines which save energy (well in the US nobody thinks about energy, except for a god given right to be consumed probably) But the market currently is dominated by the rather measly VIA CPUs which have a huge following over here in Europe (and probably Asia) Well AMD currently reacts to the trend with their own line of new fast energy savers (which we will see probably in desktop boards soon, but definitely not from via :-) )
Via currently sells boatloads of their C3 stuff, and Transmeta probably would also if their stuff was available.
So where is Intel in that game. Basically nowhere, Intel itself says this is a notebook processor only.
Some third companies already produce industrial boards because the advantages of the PM over other intel designs are huge, blazingly fast, with a rather low power consumption. But those boards cost a fortune.
But the end user market is left to VIA. What happens here is basically the same thing Intel did in the 64 bit market, which basically was handed over to AMD. And if AMD can get their act together and have several companies producing boards in the ATX format using their new low powere cores, they basically will win the slowly but rapidly emerging home server market, which currently is a hobbyists market, but in a few years will become the mass market.
My neighboor spent 900 on a brand new dell dimension 2400, when i showed him my ssytem (AMD inside) that costed 750 dollars and blew his system away in terms of performance and upgradability (the dimension doesnt even have an agp slot) he immediately asked me to sell his Dell on ebay and help him build a new computer using AMD...
i think most people just don't know enough about the nuts and bolts of computer hardware to make an informed decision with their dollars.. they see dell or maybe gateway/compaq and buy the best they can afford...