£398.24 Acer Aspire M3200, AMD Phenom 9500 Quad Core CPU, 3GB, 1TB Hard disk, HDMI! @ Ebuyer - HotUKDeals
We use cookie files to improve site functionality and personalisation. By continuing to use HotUKDeals, you accept our cookie and privacy policy.
Get the HotUKDeals app free at Google Play

Search Error

An error occurred when searching, please try again!

Login / Sign UpSubmit
206Expired

£398.24 Acer Aspire M3200, AMD Phenom 9500 Quad Core CPU, 3GB, 1TB Hard disk, HDMI! @ Ebuyer

£398.24 @ Ebuyer
The ATI Hybrid CrossFireX capable technology combined with AMD's Phenom processor give the Aspire M3200 maximum entertainment potential at an affordable price. The Aspire M3200 comes with Dolby home… Read More
siliconbits Avatar
8y, 6m agoFound 8 years, 6 months ago
The ATI Hybrid CrossFireX capable technology combined with AMD's Phenom processor give the Aspire M3200 maximum entertainment potential at an affordable price.
The Aspire M3200 comes with Dolby home theater virtual surround sound which offers a superior audio experience.
Combined with multi-in-one card reader, 7.1-channel audio support and full TV access through the PC, this desktop is your true digital home centerpiece.
Acers Aspire M3200 is designed to maximise the use of ATIs Hybrid Crossfire X technology which utilises an additional VGA card on a single motherboard.
By combining two integrated graphics processors, it offers up to a 70 percent improvement in graphics versus the standard motherboard with a single VGA card.
The M3200 comes with a Quad Core AMD Processor with 512kb, running at 2.2GHz; it is complemented by 3GB memory, a DVD writer and two 500GB Western Digital hard disk drives.
Surprisingly, Acer ships the M3200 with a Geforce 8400 video card which can be removed in order to use the onboard HDMI.
Amongst other added value speciaifcations, there's a Firewire port, Windows XP Home Premium, a card reader and Microsoft Works 8.5
More From Ebuyer:

All Comments

(29) Jump to unreadPost a comment
Comments/page:
#1
This anygood?

Will this run mkv (HD) files?,
#2
Looks quite good for the price...Any puter runs MKV files. I have my Pent.4 hooked up to my TV with s-video. Most files I play are HD quality, all be it not on an HD TV. I think you will need to check the spec of the graphics card to see if it exports HD quality with the correct leads from the card to your HD TV. Other than that, it will handle most of which you throw at it. Upgrade the card for a great puter and sell the 8400 on evilbay :thumbsup:
#3
Yeah , my pc goes into meltdown when trying to play high def files!
#4
rehydrated
Looks quite good for the price...Any puter runs MKV files. I have my Pent.4 hooked up to my TV with s-video. Most files I play are HD quality, all be it not on an HD TV. I think you will need to check the spec of the graphics card to see if it exports HD quality with the correct leads from the card to your HD TV. Other than that, it will handle most of which you throw at it. Upgrade the card for a great puter and sell the 8400 on evilbay :thumbsup:


Get an ATI card and xfire it
#5
Looks pretty tidy for the price, shame about having to remove the graphics card to get the on-board HDMI working, but still hot from me.
#6
Hmmm not sure about the AMD processor, I've always preferred Intel.
#7
I used to be up to speed on I.T kit, but over the last few years I'm far less so.

But, having said that, for processors it's all about cache, and this processor only has 512kb per processor by the sounds of it.
The entry level intel quad-core is faster (2.4ghz) and has 8mb cache...which is vastly more.

Now back in the day, Athlon used to always have less cache than Intel and used to somehow make up for (cue techs) but that was 512kb compared to 1mb, not 512kb compared to 8mb.

The graphics card on this is lousy as well - and bizarrely an Nvidia on an ATI Crossfire motherboard.

You could build the samee system with a really good graphics card for a similar price - or go to the likes of Dell Outlet for a refurb XPS 420.
The only thing that is particularly good on this is the 1Tb of disk space - but it's not a bad deal, just not a particularly good one.

Cue tek heads who will prove how the Phenom knocks the Intel quad-core into a cocked-hat.
#8
ourdave
I used to be up to speed on I.T kit, but over the last few years I'm far less so.

But, having said that, for processors it's all about cache, and this processor only has 512kb per processor by the sounds of it.
The entry level intel quad-core is faster (2.4ghz) and has 8mb cache...which is vastly more.

Now back in the day, Athlon used to always have less cache than Intel and used to somehow make up for (cue techs) but that was 512kb compared to 1mb, not 512kb compared to 8mb.

The graphics card on this is lousy as well - and bizarrely an Nvidia on an ATI Crossfire motherboard.

You could build the samee system with a really good graphics card for a similar price - or go to the likes of Dell Outlet for a refurb XPS 420.
The only thing that is particularly good on this is the 1Tb of disk space - but it's not a bad deal, just not a particularly good one.

Cue tek heads who will prove how the Phenom knocks the Intel quad-core into a cocked-hat.


Dave... Link welcomed...
#9
mamun92
Hmmm not sure about the AMD processor, I've always preferred Intel.


There's absolutely nothing wrong with AMD processors. They are not as fast as Intel counterpart but they are pretty price competitive. Back in the era when Pentium 4 came out, AMD Athlons were faster, emit less heat, cheaper and more power efficient. Back then people who bought Pentium 4s either had no brains or average Joe knowing only the Intel brand. It was the pressure that AMD put to Intel helped Intel to ditch it's poor Pentium 4 infamous Netburst architecture to bring the Core architecture. It's a shame that their R&D funding, being 1/10th of Intel, cannot keep up the competitive edge this round. Phenoms are still good processors, just not as good as Core2.

AMD is very competitive on the server market, which requires reliability more than anything. AMD and Intel are both first tier manufactures, and there's nothing to worry about.

Windows do not support having multiple graphics card of different vendors (Nvidia and ATI) hence the need to remove the GeForce card to use HDMI.
#10
The Phenom isn't as good as the Qxxxx's, but Cache, Clock Speed or any other number has nothing to do with it - it's all about architecture baby.

I'm running 2x nVidia cards on a crossfire board - your point?

The 8400GS isn't a bad card for non-gaming stuff.

£360 for all this is a bargain
#11
ourdave

But, having said that, for processors it's all about cache, and this processor only has 512kb per processor by the sounds of it.
The entry level intel quad-core is faster (2.4ghz) and has 8mb cache...which is vastly more.


Processors are not about caches but architecture.
I an write a whole report on why Intel need so much cache while AMD can do with less.
Basically AMD's built-in memory controller on the CPU had eliminate the need of large cache, which AMD cannot afford because of her smaller company size and capital for manufacturing capacity.
Intel Cores do not have on-board memory controllers. Which means CPU relies on the northbridge to pull data of RAM, which means the memory access is slower, and require larger caches as buffer zones. Intel can afford this because of her larger manufacturing capacity to make CPU with huge caches.

Cache size is like the MHz war, bigger is not necessarily better. Pentium 4 3Ghz with double the cache, is 20% slower than a 2.2Ghz Athlon with half the cache. Architecture matters much more than marketing terms.

This is to be changed with the next gen Corei7s though. Intel is packing on-board memory controllers on the next gen CPUs.
#12
AnnihilatorX
Windows do not support having multiple graphics card of different vendors (Nvidia and ATI) hence the need to remove the GeForce card to use HDMI.


No that is not right. The problem is that an inbuilt ATI graphics can only work in conjunction with an external card from the same manufacturer (Surroundview).
#13
Edit on that last post - 512kb vs 2mb.....times it by 4 = 2mb total for the Phenom, and 8mb total for the Intel.

Just seen a thread on Hexus where the Intel actually knocks the AMD Phenom into that cocked hat. So, stick with Intel.
#14
AnnihilatorX
Processors are not about caches but architecture.
I an write a whole report on why Intel need so much cache while AMD can do with less.
Basically AMD's built-in memory controller on the CPU had eliminate the need of large cache, which AMD cannot afford because of her smaller company size and capital for manufacturing capacity.
Intel Cores do not have on-board memory controllers. Which means CPU relies on the northbridge to pull data of RAM, which means the memory access is slower, and require larger caches as buffer zones. Intel can afford this because of her larger manufacturing capacity to make CPU with huge caches.

Cache size is like the MHz war, bigger is not necessarily better. Pentium 4 3Ghz with double the cache, is 20% slower than a 2.2Ghz Athlon with half the cache. Architecture matters much more than marketing terms.

This is to be changed with the next gen Corei7s though. Intel is packing on-board memory controllers on the next gen CPUs.


Partly true. AMD have lower cache memory to reduce the cost of their processors. Cache memory, as you know, is expensive. Intel have the lead in architecture due to shared cache memory across all cores, more instructions per clock cycle and smaller branch stack. There are other innovative technologies of the Core 2 that gives the speed increase. You are correct about the memory controllers.
#15
Blimey, that didn't take long.

Ta very much from the techies for explaining it all.

For someone asking for a link - I've just built a PC from a combination of Dabs and Ebuyer with Coolermaster case and PSU, decent motherboard, quad-core 2400, some fast memory, 8800GT graphics, dvd writer, card reader, hard disks and what not - for around £330. Bigger hard disks and the price of some OEM Vista software would take it to around the £400 mark that this pc is at.

Like I said, this isn't a poor deal at all, you can just do better.
#16
ElliottC
No that is not right. The problem is that an inbuilt ATI graphics can only work in conjunction with an external card from the same manufacturer (Surroundview).


I believe that the onboard graphics card will work by itself.
But you are misunderstanding me. I am not talking about CrossFire.

I should have said Windows Vista does not support in any system ATI and Nvidia cards together. One of them would not work and would be deactivated. This is because Vista does not support loading of 2 graphic card drivers from 2 different vendors.

Normally, if you have 2 different graphic cards, you can either Crossfire/SLI them, or use them separately with one extending the desktop to dual display. If it is not for the fact that the Nvidia card is not ATI, the HDMI of the built-in onboard GPU would have worked. That's a Windows Vista driver limitation issue, not technical issue.

EDIT:
My bad. It wouldn't actually make any difference as you would still need to remove the graphic card. But I can see this configuration working with Windows XP, depending on BIOS settings available to select first display device.
#17
are you sure? I know someone who has SLi'd nVidia cards and then a PCI ATi card for extra monitors...

EDIT - D'oh, he has GTX260's and an 8400GS - aka 3x nVidia cards...
#18
AnnihilatorX
I believe that the onboard graphics card will work by itself.
But you are misunderstanding me. I am not talking about CrossFire.

I should have said Windows Vista does not support in any system ATI and Nvidia cards together. One of them would not work and would be deactivated. This is because Vista does not support loading of 2 graphic card drivers from 2 different vendors.

Normally, if you have 2 different graphic cards, you can either Crossfire/SLI them, or use them separately with one extending the desktop to dual display. If it is not for the fact that the Nvidia card is not ATI, the HDMI of the built-in onboard GPU would have worked. That's a Windows Vista driver limitation issue, not technical issue.


The onboard HDMI and nvidia is a technical issue not a Windows issue. ATI drivers will support the SurroundView functionality which is built into the ATI graphics driver but it stands to reason (or maybe unreasonable in some eyes) that nVidia will not enable SurroundView and vice versa.

ATI and nVidia cards can be used together by the way but the onboard ATI graphics cannot be used with an external nVidia graphics card (see http://www.geeks3d.com/?p=1232). One thing to bear in mind is that only one driver is required because the other graphics card will be supported by Vista under a generic driver (because ATI and nVidia drivers will conflict as you mentioned). It is also worth noting that doing this will also disable the Aero interface.
#19
My experience with Intel Vs AMD quad-core chips is that Intel is definitely quite a lot quicker single or even dual core, but if you have an application which can actually use all 4 cores then the AMD chips can catch and even overtake up for very memory intensive applications. That's because the Intel FSB design can be quite a bottleneck for applications which require a lot of memory bandwidth.

The FSB performance limit is why Intel are moving to QuickPath on their next generation "Nehalem"/Core i7 architecture - essentially ripping off AMD's much better hypertransport design.

As most people won't have applications which are very memory bandwidth intensive and use all 4 cores I'd have thought that the current Intel designs would perform better for most. Mind you, at under £400 it's hard to complain about the sort of performance you would expect to get from this machine, so I'm voting hot!
#20
Spod
My experience with Intel Vs AMD quad-core chips is that Intel is definitely quite a lot quicker single or even dual core, but if you have an application which can actually use all 4 cores then the AMD chips can catch and even overtake up for very memory intensive applications. That's because the Intel FSB design can be quite a bottleneck for applications which require a lot of memory bandwidth.

The FSB performance limit is why Intel are moving to QuickPath on their next generation "Nehalem"/Core i7 architecture - essentially ripping off AMD's much better hypertransport design.

As most people won't have applications which are very memory bandwidth intensive and use all 4 cores I'd have thought that the current Intel designs would perform better for most. Mind you, at under £400 it's hard to complain about the sort of performance you would expect to get from this machine, so I'm voting hot!



I don't know where you get that information from. Most applications use all 4 cores, everytime a thread is spawned then an available core will be used - it is how Windows manage multi-tasking. The usage of cores for processing has no relevance to memory access. Yes, AMD have inbuilt memory controllers but for everything else, including multitasking, the inbuilt memory controller is outweighed (by a long way) mainly due to more instructions per clock cycle. Furthermore, during true multi-tasking (rather than time-slicing), the speed from the extra instructions per clock cycle means the benefit is even greater and I have yet to refer to the Core 2's reduced predictive branch stack, the greater L2 cache and a more improved multimedia instruction set.

You also mention Intel's FSB design? The FSB communication is governed by the chipset and the BIOS that provides the means to drive the communications.

I think what you are trying to refer to (in a very disjointed manner) is that applications where there is a deal of memory I/O involved (as opposed to memory intensive which is different) may show that the AMD processors being quicker for this purpose alone. This has absolutely nothing to do with the number of cores being used. Furthermore, one mustn't forget that no application has the sole purpose of reading and writing to memory.
#21
Anyone got any popcorn? This is fascinating! I've learnt more here than I ever did at Uni.


[munch, munch munch...]
#22
jaybizzle
This anygood?

Will this run mkv (HD) files?,


ofcourse it will, as long as you download a decent playr like VLC or Zoom player.
i have an old pentium 4 with 1 gb mem and a bog standard onboard graphics card and it runs perfectly onto my 37inch LCD.
This is a well hot deal.
#23
How about we just look at the benchmarks and let them decide:

http://www.cpubenchmark.net/common_cpus.html

Oh look, this processor beats an Intel Core2Duo 3GHz.
Jaybizzle, who's PC goes into meltdown, probably has something along the lines of a Pentium 4 2GHz. This processor is almost 8 times more powerful and will handle anything you throw at it.

This is a fantastic price in my opinion.
#24
Thewyzewun
How about we just look at the benchmarks and let them decide:

http://www.cpubenchmark.net/common_cpus.html

Oh look, this processor beats an Intel Core2Duo 3GHz.
Jaybizzle, who's PC goes into meltdown, probably has something along the lines of a Pentium 4 2GHz. This processor is almost 8 times more powerful and will handle anything you throw at it.

This is a fantastic price in my opinion.


Oh look, you are comparing a Core 2 Duo to a Phenom 9500. As you can see the dual core E8400 in Passmark's tests run the quad Core Phenom 9500 close - a testament to Intel's Core 2 power. However, in my tests but I actually have the E8400 ahead - and by a very signifcant margin. Furthermore, the link posted has the Q6600 ahead of the E8400 although I still have the E8400 ahead of the Q6600 which may show that the Passmark tests are heavily optimised for multi threading

The price of the computer in question is a reasonable price but even the aging Q6600 is a similarly priced yet faster processor.

More benchmarks here:

http://www.tomshardware.co.uk/charts/desktop-cpu-charts-q3-2008/benchmarks-1,31.html

And in particular, this sub-link which gives the Vantage benchmarks:

http://www.tomshardware.co.uk/charts/desktop-cpu-charts-q3-2008/PCMark-Vantage-PC-Mark-Suite,812.html

This shows that PC Mark Vantage is not congruent with the Passmark test you provided. I suspect the Passmark test use heavily multi threaded applications to produce the results which is why the aging quad core Q6600 (which is similarly priced to the Phenom 9500) is so high on the list. PC Mark Vantage does account for multi threading but not heavily optimised multi threaded applications which is why you see the E8500 trouncing the Q6600 and more importantly even the E8400 appears to outperform the Phenom 9500.

My PC Mark tests also has the E8400 ahead of the Phenom 9500 although it disagrees with the Toms Hardware results (using the same PC Mark results) in that my tests indicates that it also outperforms the Q6600, contrary to the identical test in Toms Hardware site.
#25
Fair point Elliot, no one benchmark can represent a processor's capabilities, much akin to how the Windows Score (or whatever that feature in Vista is called) can only give a vague idea of a computer's capabilities.
#26
ElliottC

I don't know where you get that information from. Most applications use all 4 cores, everytime a thread is spawned then an available core will be used - it is how Windows manage multi-tasking. The usage of cores for processing has no relevance to memory access. Yes, AMD have inbuilt memory controllers but for everything else, including multitasking, the inbuilt memory controller is outweighed (by a long way) mainly due to more instructions per clock cycle. Furthermore, during true multi-tasking (rather than time-slicing), the speed from the extra instructions per clock cycle means the benefit is even greater and I have yet to refer to the Core 2's reduced predictive branch stack, the greater L2 cache and a more improved multimedia instruction set.

You also mention Intel's FSB design? The FSB communication is governed by the chipset and the BIOS that provides the means to drive the communications.

I think what you are trying to refer to (in a very disjointed manner) is that applications where there is a deal of memory I/O involved (as opposed to memory intensive which is different) may show that the AMD processors being quicker for this purpose alone. This has absolutely nothing to do with the number of cores being used. Furthermore, one mustn't forget that no application has the sole purpose of reading and writing to memory.


Yes, I was a bit disjointed there - but the applications I use are both memory intensive and they also use high amounts of memory I/O. The memory bandwidth is important to me.

Actually there are still a hell of a lot of programs which don't make good use of multiple threads. Quite a few do but you'd be surprised just how many still do not. Yes, Windows is, by nature multi-threaded, but apart from a few top-end games and applications, very few single programs manage to get all 4 cores running at anything close to 100% CPU activity. (Apart from Norton bloatware of course! :lol: )

I work with a finite element analysis simulation program on twin-processor clusters (a very unusual type of program but it does push the limits of both CPU and memory). On an Intel twin-processor Xeon (4-cores per chip) machine, with a 1600 FSB, the program scales well to 4 cores, gets a bit more performance out of cores 5 and 6 but can't make any use of the 7th and 8th cores because it saturates memory bandwidth. On dual-processor quad-core Opteron processors the program is definitely slower on 1 core but scales well all the way through to all 8 cores. When using all 8 cores the Opteron does outperform the Intel machine. (Both use similar speed DDR2 memory).

I do have access to one of the new beta Nehalem-EP machines and that simply blows the older Xeon and Opteron quad-cores out of the water in terms of both raw CPU floating-point performance and memory bandwidth. Using 2 processors and 8 cores it is running over twice as fast as anything else that I have tried to date.

Now I will be the first to admit that the sorts of simulation programs that I use are very unusual, but I have seen first hand that it is quite possible to saturate the FSB architecture when using multiple cores.

Actually, just multi-threading isn't enough to make best use of today's multi-core CPUs. If you are really going to make the best of the architecture you have to do a lot of work to make sure that you get high cache hit rates (our tests show that double-precision floating point performance is 10x as fast when you are in cache) and that you don't stall the processor - a cache line stall can cost you a lot in performance.

Although memory size is no longer a problem, memory access patterns can make a huge difference in program performance. It seems ironic that we are having to return to old techniques such as memory stripmining to make best use of the latest memory architectures. With mixed GPU/CPU architectures like Larrabee on the horizon we may even need to revert to the old Cray-style vectorisation techniques to make best use of hugely-parallel CPUs!
#27
How does this system compare with ACER Aspire M3640 from Laskys which is posted here on the same day ( http://www.hotukdeals.com/item/261478/acer-aspire-m3640-base-unit-quad-co/ ). I know hard drive is much bigger (I dont need that much anyway) and its a different processor and possibly added keyboard and mouse. Is it worth spending £50 extra for this machine? since this money could go towards buying a decent monitor. Some advice will be appreciated. Thanks.
#28
Spod
Yes, I was a bit disjointed there - but the applications I use are both memory intensive and they also use high amounts of memory I/O. The memory bandwidth is important to me.

Actually there are still a hell of a lot of programs which don't make good use of multiple threads. Quite a few do but you'd be surprised just how many still do not. Yes, Windows is, by nature multi-threaded, but apart from a few top-end games and applications, very few single programs manage to get all 4 cores running at anything close to 100% CPU activity. (Apart from Norton bloatware of course! :lol: )

I work with a finite element analysis simulation program on twin-processor clusters (a very unusual type of program but it does push the limits of both CPU and memory). On an Intel twin-processor Xeon (4-cores per chip) machine, with a 1600 FSB, the program scales well to 4 cores, gets a bit more performance out of cores 5 and 6 but can't make any use of the 7th and 8th cores because it saturates memory bandwidth. On dual-processor quad-core Opteron processors the program is definitely slower on 1 core but scales well all the way through to all 8 cores. When using all 8 cores the Opteron does outperform the Intel machine. (Both use similar speed DDR2 memory).

I do have access to one of the new beta Nehalem-EP machines and that simply blows the older Xeon and Opteron quad-cores out of the water in terms of both raw CPU floating-point performance and memory bandwidth. Using 2 processors and 8 cores it is running over twice as fast as anything else that I have tried to date.

Now I will be the first to admit that the sorts of simulation programs that I use are very unusual, but I have seen first hand that it is quite possible to saturate the FSB architecture when using multiple cores.

Actually, just multi-threading isn't enough to make best use of today's multi-core CPUs. If you are really going to make the best of the architecture you have to do a lot of work to make sure that you get high cache hit rates (our tests show that double-precision floating point performance is 10x as fast when you are in cache) and that you don't stall the processor - a cache line stall can cost you a lot in performance.

Although memory size is no longer a problem, memory access patterns can make a huge difference in program performance. It seems ironic that we are having to return to old techniques such as memory stripmining to make best use of the latest memory architectures. With mixed GPU/CPU architectures like Larrabee on the horizon we may even need to revert to the old Cray-style vectorisation techniques to make best use of hugely-parallel CPUs!


That is a much better constructed response and a pleasure to read!

Re - Number of programs that are multi-threaded. There are not many that are designed in such a way that multi-threading is optimised across all the cores but by the very nature of using Windows API they will still be multi threaded nonetheless. If you examine the number processes in Task Manager and view the number of threads per process, you can see there are many applications that have many threads - mainly due to being Windows applications using the Windows messaging and partly they spawn threads of their own. I do acknowledge your wording that many do not make "good use of multiple threads" but it is important that one realises that more cores in a processor does result in greater speed even more applications that are not optimised to take advantage of all cores. This brings me back to the earlier post regarding the rather curious comparison between an Intel E8400 and Phenom 9500 because despite my comments, the dual Core E8400 appears to be a match for the Quad Core 9500 even in benchmarks that appear to be heavily optimised for multiple cores (and in Toms Hardware tests, even outperforms it quite significantly).
#29
ahelast
How does this system compare with ACER Aspire M3640 from Laskys which is posted here on the same day. I know hard drive is much bigger (I dont need that much anyway) and its a different processor and possibly added keyboard and mouse. Is it worth spending £50 extra for this machine? since this money could go towards buying a decent monitor. Some advice will be appreciated. Thanks.



The M3640 has an older processor Q6600 which will is being phased out (although a lot later than anticipated due to its popularity). However, the Q6600 is much faster than the Phenom 9500 (the explanation behind this are in previous postings on this thread although I failed to add that Intel processors also employ shared caching across all cores rather than a fixed amount per core resulting in more speed improvement).

I don't know what graphics chipset is used so it is difficult to compare the performance of gaming, HD playback or other graphically intensive applications.

Post a Comment

You don't need an account to leave a comment. Just enter your email address. We'll keep it private.

...OR log in with your social account

...OR comment using your social account

Top of Page
Thanks for your comment! Keep it up!
We just need to have a quick look and it will be live soon.
The community is happy to hear your opinion! Keep contributing!