Omega Owners Forum
Chat Area => General Discussion Area => Topic started by: Martin_1962 on 11 November 2008, 14:05:44
-
Motherboards
Bus speed a lot are 1333 and some are 1600 - much difference?
Dual or Quad core at the same price point?
RAM for XP - will 2GB be enough?
-
My rule of thumb is always have 2gig per CPU core, so dual core will be 4gig, quad core 8gig.
XP32 won't support more than 3gb tho.
-
Motherboards
Bus speed a lot are 1333 and some are 1600 - much difference?
Dual or Quad core at the same price point?
RAM for XP - will 2GB be enough?
1333 and 1600 will differ but if you install the same spec DDR RAM..
Quad core Cpus will have meaning depending on the software you use
ie like 3d software or games etc..
2GB for Xp in my experience really is enough.. I use 1GB..
if you want really a fast machine go for high RPM disks..Not necessary
RAID but even a 10K or 15K RPM disk will really matter..
-
My rule of thumb is always have 2gig per CPU core, so dual core will be 4gig, quad core 8gig.
XP32 won't support more than 3gb tho.
must be 4 GB I think..
-
Few games 'really' benefit from quad cores at present, although there are some. That said, quads are comortably fast enough for gaming too - only if you need extreme performance in games would a dual arguably be a better choice.....
As for RAM, as mentioned above, 32-bit OS's can only see 4GB RAM in total, but bear in mind that you will not see all that as Windows takes a fair bit for address mapping of bits and pieces - example, if your gfx card has 512MB, then that amount will be included in the 4GB that Windows sees, so you are down to 3.5GB immediately....in practice, you will see somewhere between 3 and 3.3GB available with 4GB RAM installed in a 32-bit OS.
Bus speeds are really not soooo important when looking @ 1333 against 1600 - unless you are planning to squeeze every last drop of performance out of the system. Sure, the 1600 speed will be technically faster, but whether that gain is realised in a real-world sense in day-to-day use is debatable for most people.
High RPM drives are excellent for OS installs, eg a Raptor or similar......helps with boot times, game loads and so on.....but often there is a trade off between performance and size......that said, the latest Samsung F1 range offers a bewitching combination of speed and performance......
Finally, motherboards - remember that this is the basis of the whole setup and it is never advisable to scrimp on this area. Everything else plugs into the motherboard so wise choice is prudent here.
-
No ram out there that relay fully utilises the 1600 bus speed yet.....the stuff that is compatible effectively has wait states to meet timing!
-
1333 and 1600 bus speeds will actually differ for applications that transfer huge amounts of memory chunks and if the correct spec RAM is installed or available..
But more than this if you are trying a numerical simulator on a numerical model which requires many days to work non stop you will highly benefit.. of which I was trying in university with a an old 8088 ;D ;D
-
1333 and 1600 bus speeds will actually differ for applications that transfer huge amounts of memory chunks and if the correct spec RAM is installed or available..
But more than this if you are trying a numerical simulator on a numerical model which requires many days to work non stop you will highly benefit.. of which I was trying in university with a an old 8088 ;D ;D
I would agree IF the ram technology was capable of making full use of the bus speed.....and most is not yet.
Trouble is that the figure is exactly that, the max clock rate of the ram and is not directly related to the speed of fectch/write cycles in the ram.
What many ram suppliers are currently doing is taking an extra 0.5-1 clock cycle for access......hence negating the faster bus speed!
-
1333 and 1600 bus speeds will actually differ for applications that transfer huge amounts of memory chunks and if the correct spec RAM is installed or available..
But more than this if you are trying a numerical simulator on a numerical model which requires many days to work non stop you will highly benefit.. of which I was trying in university with a an old 8088 ;D ;D
I would agree IF the ram technology was capable of making full use of the bus speed.....and most is not yet.
Trouble is that the figure is exactly that, the max clock rate of the ram and is not directly related to the speed of fectch/write cycles in the ram.
What many ram suppliers are currently doing is taking an extra 0.5-1 clock cycle for access......hence negating the faster bus speed!
Yep, age old method of faster 'headline' speed, with worse latencies ! Naughty little monkeys.....
-
1333 and 1600 bus speeds will actually differ for applications that transfer huge amounts of memory chunks and if the correct spec RAM is installed or available..
But more than this if you are trying a numerical simulator on a numerical model which requires many days to work non stop you will highly benefit.. of which I was trying in university with a an old 8088 ;D ;D
I would agree IF the ram technology was capable of making full use of the bus speed.....and most is not yet.
Trouble is that the figure is exactly that, the max clock rate of the ram and is not directly related to the speed of fectch/write cycles in the ram.
What many ram suppliers are currently doing is taking an extra 0.5-1 clock cycle for access......hence negating the faster bus speed!
yep..because producing the original spec memory unit is expensive :y
and I must add some cheap modules require more cycles to prepare data..
but if you are ready to pay a serious amount for memory modules
here is the benefit (dont know where you will find them though ;D)
(http://i181.photobucket.com/albums/x80/mecdv6/memperformance.jpg)
edit: found those in toms hardware guide
http://www.tomshardware.com/reviews/arms-race-ddr3-1800-ram,1676-10.html
http://www.tomshardware.com/reviews/arms-race-ddr3-1800-ram,1676-11.html
-
Martin Imber:
If you are going dual/quad core, seriously consider going to Vista, and updating your very old apps that may not run (yet to find any myself that are good under xp, but not Vista)
32bit OS can access 4gb RAM. With PC architecture, this is normally capped to 3.5Gb (bios shadowing, option roms, video window etc).
Windows architecture is (as std) 2gb user (per process), 2gb kernel (shared), but a boot.ini switch can change to 3g/1g. Kernel uses little memory, so much beyond 3g in 32bit Windows is pointless (ignoring non consumer, and PAE technologies in servers)
64bit windows overcomes this, each version of (consumer) windows puts artificial limits on. 64bit has trouble with lack of drivers, but Vista is again the best. Older nasty printers are probably biggest issue.
FSB speeds, RAM is slow, not worth spending a premium on faster fsb speeds, ram no longer truely sycronous. DDR2 is quad pumped, but internally cannot deliver.
I know you don't like change and will find any excuse to slate Vista, once you 'understand' it, trust me, it makes sense.
Memory amounts. 1gb min for xp, 2gb min for vista will give speedy system :)
-
My rule of thumb is always have 2gig per CPU core, so dual core will be 4gig, quad core 8gig.
XP32 won't support more than 3gb tho.
thats because linux needs it ;D
Linux memory management is awful. Actually, Linux resource management is awful (non existent) ;D
-
BTW already have case, FDD, DVD Writer, as I am rebuilding the old PC
-
I get confussed easy. When I started reading the thread I was going to mention that a lot more boards are now coming out with DDR3 (my latest board has 4x DDR2 and 2x DDR3) so therefore would it be worth considering the DDR3 againt cost of purchase and appliction/performance.
I am still running DDR2 till DDR3 prices settle.
But after seeing the chart that Cem listed I am not sure anymore. :-/
-
BTW already have case, FDD, DVD Writer, as I am rebuilding the old PC
FDD = £5, though does anyone still use
DVD Writer = £12
Case = £6
For that kind of money, give old one to kids, and build entirely new
-
I get confussed easy. When I started reading the thread I was going to mention that a lot more boards are now coming out with DDR3 (my latest board has 4x DDR2 and 2x DDR3) so therefore would it be worth considering the DDR3 againt cost of purchase and appliction/performance.
I am still running DDR2 till DDR3 prices settle.
But after seeing the chart that Cem listed I am not sure anymore. :-/
Same as dual AGP/PCIe boards, you get a compromise between the 2, rather than max performance from one.
-
My rule of thumb is always have 2gig per CPU core, so dual core will be 4gig, quad core 8gig.
XP32 won't support more than 3gb tho.
thats because linux needs it ;D
Linux memory management is awful. Actually, Linux resource management is awful (non existent) ;D
does hell ;D *never* had any memory issues on both of my laptops, desktop or server. (admittedly I did have some issues when running Debian to begin with, but that was quite some time ago)
makes more sense to match memory sizes per core than mix & match, purely on a performance level!
had 3gig in my laptop when bought, since upping to 4gig it is noticeably faster. Even then, memory usage with just 3gig there was only around 1gig with no swap during normal running, at most, 2.5gig used during compiling big apps (even then swap was hardly used)
Noticed similiar performance increases on my brothers desktop (dual core) running Vista64, he had 3gig prior, upped to 4gig.
-
My rule of thumb is always have 2gig per CPU core, so dual core will be 4gig, quad core 8gig.
XP32 won't support more than 3gb tho.
thats because linux needs it ;D
Linux memory management is awful. Actually, Linux resource management is awful (non existent) ;D
does hell ;D *never* had any memory issues on both of my laptops, desktop or server. (admittedly I did have some issues when running Debian to begin with, but that was quite some time ago)
makes more sense to match memory sizes per core than mix & match, purely on a performance level!
had 3gig in my laptop when bought, since upping to 4gig it is noticeably faster. Even then, memory usage with just 3gig there was only around 1gig with no swap during normal running, at most, 2.5gig used during compiling big apps (even then swap was hardly used)
Noticed similiar performance increases on my brothers desktop (dual core) running Vista64, he had 3gig prior, upped to 4gig.
The current poor resource management in Linux is one of the inhibiting factors for it to become more than just a poor man's webserver. I think the most we trust Linux with is DNS, servicing about 80,000 queries per second, but we've had to put more servers in since moving from proper Unix to Linux to handle the load (even though the hardware is newer and faster). We trusted it with News, but hacked too often.
On every x86/x64 architecture I have seen, there is no performance reason to match memory to cores. In fact, on any chipset I have used, there is not that level of ability.
As to using 2.5Gb running little more than a compiler, that is disgraceful ;), makes even Visual Studio look efficient on memory usage ;D