Catalyst CC a lot worse than Forceware?

inscape01

Limp Gawd
Joined
Apr 14, 2006
Messages
283
I'm having a real hard time picking between the x1800xt and 7900gt so i'm taking everything into consideration. The CCC sounds bloated because it takes a long time to load and requires .net. Forceware on the other hand is nice and streamlined. Are my views correct, or is the CCC not as bad as some people make it seem? Also is the x1800xt fan really that loud, or are people exaggerating?
 
yes they are very loud. it doesnt bother some people but the x1900xt/xtx and x1800xt have the LOUDEST factory fans. the only problem i have with ATI drivers and CCC over Nvidia is that you will use around 100mb more ram just sitting on the desktop. for instance i installed the ATI drivers and CCC with an x1300pro in my comp and i was at 270mb on the desktop. i uninstalled that and went with a 6600 and i was at 175mb. this wont concern you if you have 2 gigs of ram. other than that the CCC is no big deal.
 
Three words: ATI Tray Tools ;) I don't think you even need the classic control panel installed with this thing. Takes only around five megs of memory and will even make your bed in the morning.
 
CCC doesnt take a long time to load at all, unless you consider 1-2 seconds long. The only time it ever takes longer than that for me, is the first time. After the first, its 1-2 seconds. You can also just right click the icon, and make most any change, without having to load it up.

Or use ATi Tray Tools as already mentioned.
 
The initial CCC load is between 4 and 10 seconds, then it's substantially quicker the next time.

As for why I chose control panel with cat 6.3, it allows me to select different 3D options between OpenGL and D3D. Forceware and CCC lock you into common settings regardless of mode.
 
Thanks, the ati tray tools looks pretty cool. This is the hardest video card decision ever, I read a few old threads comparing these two and there was no clear winner.
 
inscape01 said:
Thanks, the ati tray tools looks pretty cool. This is the hardest video card decision ever, I read a few old threads comparing these two and there was no clear winner.

There isn't really. I just decided to go with an X1800XT (it came yesterday). ATI's recent Linux drivers and the card being flashable to an XT PE are what tipped me over. Price was nice too, $242 shipped (after $30 rebate) from Newegg.
 
I have an 1800xt
At 100% the fan is pretty loud.
Mine is run by temperature control so it is usually about 30% fan speed which isn't that bad.

I just have my parts out on the desk though and not in a case.. so it is probably louder for me.
 
The first time I fired up my X1900xtx I thought wow this could be a problem but useing it now for a month or so to be honest I do no event notice it anymore, maybe im looseing my hearing from it :D
I will tell you its worth it the card kicks ass :p
 
The only objection I see to the ATI Control Center is that it requires Microsoft .NET Framework to run, while Forceware runs independently without any more M$ crap installed. Even though many have the .NET Framework, IMO graphics drivers should not require such things....
 
trek554 said:
yes they are very loud. it doesnt bother some people but the x1900xt/xtx and x1800xt have the LOUDEST factory fans. the only problem i have with ATI drivers and CCC over Nvidia is that you will use around 100mb more ram just sitting on the desktop. for instance i installed the ATI drivers and CCC with an x1300pro in my comp and i was at 270mb on the desktop. i uninstalled that and went with a 6600 and i was at 175mb. this wont concern you if you have 2 gigs of ram. other than that the CCC is no big deal.

1. It uses up 100MB of virtual memory, not RAM. Big difference.
2. Unless you're planning on running an ultra-silent PC, you won't care about the fan noise.
 
Great, I think i'm set on the x1800xt now. Now the next decision is whether or not 512mb over 256mb will actually be worth it. Also, I'm pretty set on wanting a Sapphire card, but how are MSI cards? For some reason I just prefer Sapphire over MSI.
 
With many graphics cards, one can overclock a lower part (say 7900GT) to a higher level (7900GTX). This may require a hardmod (voldmodding anyone?) or a softmod (unlock 6800AGP via RivaTuner). Sorry for the nV examples but since I'm an nV Nut I don't know ATI (the last ATI card I had was a 32MB PCI Radeon VE....still in my mom's word processing box). However, what overclocking can't make up for is lack of memory. Overclocking cannot force 500MB of textures (a la Doom 3 Engine Ultra Quality) into 256MB of memory. It's really up to you. What quality do you demand? Do you like always being able to run the latest and greatest games on their highest settings? If so, the 512MB part is for you (I'm going dual 7900GTXs because I am like this). If you worry less about quality of graphics, you can get away with the 256MB card.

On Sapphire vs. MSI - From what I know, Sapphire is one of the best (if not the best) ATI shop out there...
 
PWMK2 said:
1. It uses up 100MB of virtual memory, not RAM. Big difference.
2. Unless you're planning on running an ultra-silent PC, you won't care about the fan noise.
well how is that not ram? if you have 1 gig physical ram doesnt that get used first? for a game like Fear i didnt really have to hit the hard drive with an nvidia card. i was right at 1 gig in the task manager. when i used the ati card the game hitch around corners so i looked at the task manager and it was over 1 gig. these were on the same game settings.
 
PWMK2 said:
1. It uses up 100MB of virtual memory, not RAM. Big difference.
2. Unless you're planning on running an ultra-silent PC, you won't care about the fan noise.
While there is a big difference between virtual memory and physical pages, the "Mem Usage" column of Task Manager does correspond to the size of the working set of the process - the portion of virtual address space that is currently resident in physical RAM.

This is somewhat irrelevant though because with the combination of ATI Tray Tools and ATI Tool, you don't ever need to install CCC.
 
CCC never takes more than 2-3 seconds to load for me.

I just don't like the fact that the 6.3 and 6.4 drivers just DON'T WORK.
 
I've used both of them recently.

The ATI drivers are more bloated, and try to do too much for you. I also experienced unstability, I had the ATI serivce crash on me several times over several installations of Windows. In fact, the ATI drivers got so bad that I switched to Omega; the performance difference was minimal, but the stability difference was better. The nVidia drivers are much more streamlined, and offer more tweaking options (once you install coolbits). They are much more stable, I've only had the nVidia service crash on me once or twice.

Cliffs Notes:
nVidia drivers > ATI drivers
 
NoDamage said:
While there is a big difference between virtual memory and physical pages, the "Mem Usage" column of Task Manager does correspond to the size of the working set of the process - the portion of virtual address space that is currently resident in physical RAM.
thanks for saying that. it drives me nuts when people say "it uses virtual memory" about the CCC. memory is memory until you physically run out of it and have to use the hard drive.
 
ironforge said:
I have an 1800xt
At 100% the fan is pretty loud.
Mine is run by temperature control so it is usually about 30% fan speed which isn't that bad.

I just have my parts out on the desk though and not in a case.. so it is probably louder for me.


I agree. After I saw a video of the 1900xtx fan, I thought it was insanely loud. To my surprise my brothers 1900xt wasn't loud at all unless it gets up to 100% which I've never seen it do in game. But at 100% it's definately loud.
 
trek554 said:
thanks for saying that. it drives me nuts when people say "it uses virtual memory" about the CCC. memory is memory until you physically run out of it and have to use the hard drive.
Well, that's not quite right either. Windows memory management is kind of a complicated topic to explain, but it isn't quite as simple as using the hard drive when the physical RAM runs out.

Every application uses the virtual memory abstraction because that is how the operating system was designed. For the most part, Windows constantly utilizes almost all of the available RAM in the system for various purposes, so the notion of running out of RAM and spilling over onto the hard drive doesn't really make sense in this context.

(Task Manager is very misleading in this regard, the amount of memory usage it reports doesn't really mean what most people think it means.)

Anyway, on the desktop the X1900XT fan is pretty much as quiet as most other standard video cards. It is only when you go into 3D mode and the temps climb up that the fan starts roaring, and yes it is incredibly annoying. I wouldn't want to run that card without some aftermarket cooling.
 
NoDamage said:
Well, that's not quite right either. Windows memory management is kind of a complicated topic to explain, but it isn't quite as simple as using the hard drive when the physical RAM runs out.

Every application uses the virtual memory abstraction because that is how the operating system was designed. For the most part, Windows constantly utilizes almost all of the available RAM in the system for various purposes, so the notion of running out of RAM and spilling over onto the hard drive doesn't really make sense in this context.

(Task Manager is very misleading in this regard, the amount of memory usage it reports doesn't really mean what most people think it means.)

You can select a column in the task manager to look at the virtual memory usage ;)
 
In changing from a 7900GT to this X1900XT, the biggest difference was by far the drivers. These ATI things are beyond garbage compared to NV forceware. It seriously takes longer for my system to completely finish booting now just because of the change. I uninstalled, ran driver cleaner several times to remove the NV drivers, and it still sucks. Maybe ATI Tray Tools will be better.
 
Least said:
You can select a column in the task manager to look at the virtual memory usage ;)
That column doesn't mean what most people think it means either, actually it's probably more misleading than Mem Usage itself. VM Size refers to the portion of address space a process is using that happens to be backed by the page file. But if you look at some application and see 50 MB listed under the VM Size, it doesn't mean that 50 MB of the application's memory is sitting on the hard drive.
 
Well, we can all agree on one thing - if CCC used 100MB of physical memory instead of prioritizing memory usage into virtual memory (like all of the MS Office programs and Photshop do...), there would be problems. However, I have no problems with CCC using my virtual memory, especially considering my pagefile is set at 3GB max and I've never seen it go above 1.5GB, even when running a game. Adding stuff to virtual memory will not necessarily hinder peformance, only the loadtime for CCC (hence why it takes 4-6 seconds) and the start-up / shut-down times.

I want everyone to notice their memory usage before and after installing MS Office. It's a lot worse than CCC, trust me... hell, even MS IME (input system for foreign languages) uses up like 30MB of virtual memory but I don't really care (can't care though since I need it for my Japanese homework... :D)
 
PWMK2 said:
Well, we can all agree on one thing - if CCC used 100MB of physical memory instead of prioritizing memory usage into virtual memory (like all of the MS Office programs and Photshop do...), there would be problems. However, I have no problems with CCC using my virtual memory, especially considering my pagefile is set at 3GB max and I've never seen it go above 1.5GB, even when running a game. Adding stuff to virtual memory will not necessarily hinder peformance, only the loadtime for CCC (hence why it takes 4-6 seconds) and the start-up / shut-down times.

I want everyone to notice their memory usage before and after installing MS Office. It's a lot worse than CCC, trust me... hell, even MS IME (input system for foreign languages) uses up like 30MB of virtual memory but I don't really care (can't care though since I need it for my Japanese homework... :D)
im sorry that just doesnt make any sense. how can an application just decide to use virtual memory not physical memory? i ran a little test on my other pc with Fear and 1 gig of memory. the game was right at 1 gig in the task manager and was smooth going around corners. i opened up a few windows and restarted the game and was now a little over 1 gig in the task manager. the game hitched at every corner. this makes sense since i was now using more than physical ram and was hitting the hard drive. this seems to prove to me that amount shown in the task manager is the actual amount being used. so if you have 1 gig installed in your pc and you see the usage at 1 gig you know you are almost out of physical memory. so like i said way earlier you will use up more ram with ati and ccc compared to nvidia. this 100mb WILL make a difference in some games if you only have 1 gig of memory. you can call it virtual memory all you want but the 100mb is being taken away from usable ram.
 
PWMK2 said:
Well, we can all agree on one thing - if CCC used 100MB of physical memory instead of prioritizing memory usage into virtual memory (like all of the MS Office programs and Photshop do...), there would be problems. However, I have no problems with CCC using my virtual memory, especially considering my pagefile is set at 3GB max and I've never seen it go above 1.5GB, even when running a game. Adding stuff to virtual memory will not necessarily hinder peformance, only the loadtime for CCC (hence why it takes 4-6 seconds) and the start-up / shut-down times.

I want everyone to notice their memory usage before and after installing MS Office. It's a lot worse than CCC, trust me... hell, even MS IME (input system for foreign languages) uses up like 30MB of virtual memory but I don't really care (can't care though since I need it for my Japanese homework... :D)
If CCC says Mem Usage is 100MB in Task Manager, then indeed 100MB of physical RAM is being utilized by CCC. CCC is not using your "virtual memory" in the way you think, and neither are Office or Photoshop.

Again, "virtual memory" is not simply an extension of your physical RAM onto your hard drive. If you have 1 GB of RAM and a 1 GB pagefile, it is not correct to say that you have 1 GB of physical memory and 1 GB of "virtual memory". Virtual memory is an operating systems concept that provides each process with its own virtual address space. It is implemented independently of the page file. Applications cannot prioritize memory usage into virtual memory in the way you suggest.
 
NoDamage said:
If CCC says Mem Usage is 100MB in Task Manager, then indeed 100MB of physical RAM is being utilized by CCC. CCC is not using your "virtual memory" in the way you think, and neither are Office or Photoshop.

Again, "virtual memory" is not simply an extension of your physical RAM onto your hard drive. If you have 1 GB of RAM and a 1 GB pagefile, it is not correct to say that you have 1 GB of physical memory and 1 GB of "virtual memory". Virtual memory is an operating systems concept that provides each process with its own virtual address space. It is implemented independently of the page file. Applications cannot prioritize memory usage into virtual memory in the way you suggest.
thank you. i knew i wasnt completely crazy. where did he get that crazy idea from? hopefully what i said in my last post made a little sense too. ;)
 
Where did I get that crazy idea from?

1. Get a monitoring program like Motherboard monitor.
2. Look at your physical memory usage before and after installing CCC.
3. Notice how they're they're almost exactly the same (give or take 20 MB).
4. Notice how virtual memory has gone up 100 MB.
5. See how crazy of an idea it is.

I have not once used task manager to monitor my memory usage. But if you DID, you would notice that the three instances of CLI.EXE never go above 15MB total.

If CCC used up 100MB of physical memory, then you would see a huge difference in scores in benchmark programs (and in FPS counts in games) between systems running ATITool and systems running CCC -- which you don't really. I'm not defending CCC here necessarily, I'm just saying that it is NOT using up 100MB of physical memory. There are tons of other good reasons to install ATITool over CCC (including MUCH better OCing and monitoring options), but performance ain't one of them.

And yes, you can prioritize memory usage into virtual memory. Windows does this all the time. This is why the pagefile has a minimum size in the first place. Windows can grant virtual memory addresses to any application that requests it. As an example, my computer has 1024MB of virtual memory. When I start up Windows, according to Motherboard Monitor, I have 486MB of physical memory allocated. But for the pagefile, I have 412MB of virtual memory allocated (it was 312 before I had CCC installed). Hmm... 486 + 412 = 898. I have more than 898MB of RAM. Why is the pagefile being accessed, then? Because certain components of Windows, as well as other programs, are put into the pagefile when they're not being used and are then loaded into physical memory. CCC is one of these programs that does this.
 
First off, I LOVE my X1900 Crossfire setup when I'm actually gaming.

But they've made the process of setting up a new overclocked rig a freakin' nightmare. Why? Not because they crash, not because they add a lot of heat to the loop, not because they give my 530w PSU a workout, and not because they're noisy.

It's because CCC adds FORTY FIVE SECONDS to my Windows XP load times...that really adds up when you're rebooting every few minutes to test a change, or crashing because you've pushed something too far, or just the restarts due to other software installs.

Seriously...45 seconds at XP load, AFTER getting logged on, with 50%-100% CPU load as CCC goes through its thing (and why does CCC have to hit my floppy drive?). This is with 2gb RAM, it's even worse with 1gb.

Hell, I've got an ancient HP 700mhz 256mb Celeron laptop that loads XP faster, and that's including loading all the HP power management apps, spyware and antivirus, HP update, wireless PCMIA card drivers, Bluetooth card drivers, etc.

A nVidia PCI-e 6200, or PCI MX card adds nothing to the load time overhead, ten seconds after XP login I'm fully stable and the only activity is XP doing a little diddling with the pagefile. Or even a single X1900 using ATI's older CP instead of CCC.

Unfortunately, I have NO choice in the matter, I don't have the option of using the old leaner meaner ATI CP, or ATI Tray Tools, etc, because the only way Crossfire will work is with CCC running.

It's not a big deal once you've got your system dialed in and aren't rebooting on a regular basis, but it still speaks volumes as to how bloated CCC is.
 
PWMK2 said:
Where did I get that crazy idea from?

1. Get a monitoring program like Motherboard monitor.
2. Look at your physical memory usage before and after installing CCC.
3. Notice how they're they're almost exactly the same (give or take 20 MB).
4. Notice how virtual memory has gone up 100 MB.
5. See how crazy of an idea it is.

I have not once used task manager to monitor my memory usage. But if you DID, you would notice that the three instances of CLI.EXE never go above 15MB total.

If CCC used up 100MB of physical memory, then you would see a huge difference in scores in benchmark programs (and in FPS counts in games) between systems running ATITool and systems running CCC -- which you don't really. I'm not defending CCC here necessarily, I'm just saying that it is NOT using up 100MB of physical memory. There are tons of other good reasons to install ATITool over CCC (including MUCH better OCing and monitoring options), but performance ain't one of them.

And yes, you can prioritize memory usage into virtual memory. Windows does this all the time. This is why the pagefile has a minimum size in the first place. Windows can grant virtual memory addresses to any application that requests it. As an example, my computer has 1024MB of virtual memory. When I start up Windows, according to Motherboard Monitor, I have 486MB of physical memory allocated. But for the pagefile, I have 412MB of virtual memory allocated (it was 312 before I had CCC installed). Hmm... 486 + 412 = 898. I have more than 898MB of RAM. Why is the pagefile being accessed, then? Because certain components of Windows, as well as other programs, are put into the pagefile when they're not being used and are then loaded into physical memory. CCC is one of these programs that does this.
to be honest this is getting confusing. you can still call it virtual memory but does take away from available ram. the little test i did with Fear was the same results i got with an Ati card and CCC using 100mb of ram. basically what i am saying is that 100mb is used up so if you are playing a game that uses right at a gig with an nvidia you will be over 1 gig of memory usage with Ati and the CCC. if you have 1 gig of physical ram you will hitch a little when a game requires more than that. WITH ATI AND CCC INSTALLED YOU WILL HAVE 100MB LESS MEMORY TO UTILISE.
 
trek554 said:
to be honest this is getting confusing. you can still call it virtual memory but does take away from available ram. the little test i did with Fear was the same results i got with an Ati card and CCC using 100mb of ram. basically what i am saying is that 100mb is used up so if you are playing a game that uses right at a gig with an nvidia you will be over 1 gig of memory usage with Ati and the CCC. if you have 1 gig of physical ram you will hitch a little when a game requires more than that. WITH ATI AND CCC INSTALLED YOU WILL HAVE 100MB LESS MEMORY TO UTILISE.

Thats just plain wrong. There is a difference between physical memory, which is your sticks of ram, and virtual memory, which is a data cache on your hard disk. Windows uses the virtual memory when it runs out of physical ram. The amount of virtual memory changes constantly as more and more programs are opened until it reaches the amount you have set in system properties, unless you let windows manage the size.

So you dont lose 100mb physical memory at all. Windows has just allocated another 100mb of hard drive space ready for if it runs out of physical memory. CCC totals only about 15 or so meg of physical memory on my system. With my pagefile at 518 meg
 
Skirrow said:
Thats just plain wrong. There is a difference between physical memory, which is your sticks of ram, and virtual memory, which is a data cache on your hard disk. Windows uses the virtual memory when it runs out of physical ram. The amount of virtual memory changes constantly as more and more programs are opened until it reaches the amount you have set in system properties, unless you let windows manage the size.

So you dont lose 100mb physical memory at all. Windows has just allocated another 100mb of hard drive space ready for if it runs out of physical memory. CCC totals only about 15 or so meg of physical memory on my system. With my pagefile at 518 meg
THATS EXACTLY WHAT I JUST SAID. you must be misreading or maybe i wasnt clear. you run out of physical ram to use 100mb sooner with Ati and CCC compared to nvidia. once out of physical ram then you will be hitting the hard drive for help.
 
so just increase the size of your page file then.

It's not like ccc will start by itself when you are gaming and be moved from virtual memory to your "real" memory.

when you are gaming ccc will take up about 15mb, the rest will be tucked away on your harddrive.
 
Spank said:
so just increase the size of your page file then.

It's not like ccc will start by itself when you are gaming and be moved from virtual memory to your "real" memory.

when you are gaming ccc will take up about 15mb, the rest will be tucked away on your harddrive.
i still dont see how its not using "real" memory. i am not going to repeat everything i have already said. i have done my own testing and know what effect the ccc has on memory. maybe you guys know more than me but i know what results i have seen with my own eyes. ;)
 
If you're claiming CCC adds 45 seconds to your load time, then you are mistaken. Thats just pure lack of knowledge on your part.
 
PWMK2 said:
And yes, you can prioritize memory usage into virtual memory. Windows does this all the time. This is why the pagefile has a minimum size in the first place. Windows can grant virtual memory addresses to any application that requests it. As an example, my computer has 1024MB of virtual memory. When I start up Windows, according to Motherboard Monitor, I have 486MB of physical memory allocated. But for the pagefile, I have 412MB of virtual memory allocated (it was 312 before I had CCC installed). Hmm... 486 + 412 = 898. I have more than 898MB of RAM. Why is the pagefile being accessed, then? Because certain components of Windows, as well as other programs, are put into the pagefile when they're not being used and are then loaded into physical memory. CCC is one of these programs that does this.
Skirrow said:
Thats just plain wrong. There is a difference between physical memory, which is your sticks of ram, and virtual memory, which is a data cache on your hard disk. Windows uses the virtual memory when it runs out of physical ram. The amount of virtual memory changes constantly as more and more programs are opened until it reaches the amount you have set in system properties, unless you let windows manage the size.

So you dont lose 100mb physical memory at all. Windows has just allocated another 100mb of hard drive space ready for if it runs out of physical memory. CCC totals only about 15 or so meg of physical memory on my system. With my pagefile at 518 meg
Both of you do not understand what virtual memory actually is. Like I already said, calling the contents of the page file "virtual memory" that is utilized in lieu of physical memory is a completely incorrect description. "Virtual memory" is not a data cache on your hard disk to be utilized when Windows runs out of physical RAM.

Virtual memory is the term for memory management systems that utilize virtual memory addressing, providing a virtual address space of 4 GB (on 32-bit systems) for each process. This has nothing to do with the page file or the hard disk. Every application on the system utilizes its own virtual address space, and hence uses virtual memory.

The notion of the operating system granting virtual memory addresses to individual applications does not make any sense and completely misunderstands virtual memory. Every application uses a virtual address space with virtual memory addressing already. It is the operating system and hardware that map those virtual memory addresses onto physical addresses in RAM. This allows multiple processes to share the same physical memory while allowing for flexible allocation (a contiguous block of virtual address space does not have to correspond to a contiguous block of physical ram) and security (each process's virtual address space is isolated, so one process can't overwrite another process's memory.)

Now, paging is a common method by which virtual memory systems are actually implemented. With paging, memory is divided into separate chunks (typically 4 kB) and all memory access operations are performed in page sized units. Note that even with paging, the notion of a page file or swap file or hard drive is still not quite relevant yet. A system which implements paging can do so without using a page file.

So, what is the page file and how is it relevant to the above? The page file is used to provide backing for certain pages that are resident in memory. Your physical memory is divided into pages, and each page corresponds to a page of virtual address space of a different process. Now, obviously, at some point you will run out of physical memory, and so a page will need to be removed from physical RAM. Some types of pages will be stored in the page file when they are kicked out of RAM, and this is called being backed by the page file. But just because something is backed by the page file does not mean that it actually current exists in the page file.

In general, the operating system controls which pages of a given process are currently resident in memory (known as the "working set" of the process). An application cannot say "give me 50 MB of memory but put it on the hard disk" because the memory management system does not work like that, and in the context of how it actually works, that request makes absolutely no sense whatsoever. Pages will be kicked out when the operating system needs to make room for new pages, but typically the application has no control over this at all. If the operating system needs to kick your pages out of RAM, well, too bad.(There is a lot more complexity here that I am neglecting to cover as well. Pages can be trimmed out of a process's current working set by the operating system yet still be resident in RAM. And this is typically the case)

Furthermore, the statement "my computer has 1024 MB of virtual memory" does not make any sense either. While you may have a 1024 MB page file, the page file does not equal virtual memory in any sensible way.

And finally, the numbers that Task Manager and most other software monitoring programs report for memory usage, do not actually correspond to the full memory usage of the system. For instance, the Commit Charge in Task Manager only reports the portion of virtual address space utilized by the system that is backed by the page file. This is nowhere near the total amount of memory in use. Software monitoring programs typically retrieve the same numbers from the operating system and are equally misleading. So the numbers you are getting from Motherboard Monitor really don't mean what you think they mean.

The numbers are especially misleading when it comes to page file usage. What Task Manager reports as VM Size is actually the "Private Bytes" of the process. If I look at a program and see a VM Size of 50 MB, this DOES NOT mean that there is 50 MB of data in the page file of that program's memory. It simply means that 50 MB of the program's memory is backed by the page file, meaning that if that 50 MB were to be kicked out of RAM, it would be put into the page file.

On a typical system with 1 GB+ RAM, your page file is being utilized much less than you think it is. The only accurate way to truly measure your page file usage is with pstat.exe off the Windows XP Support Tools package, which will give a sample output that looks something like this:
PageFile: \??\D:\pagefile.sys
Current Size: 1569792 kb Total Used: 36492 kb Peak Used 44900 kb

Memory:1047276K Avail: 580132K TotalWs: 402624K InRam Kernel: 108K P:49616K
Commit: 346536K/ 216164K Limit:2517016K Peak: 387388K Pool N:29308K P:49940K
In this example on a system with 1 GB of RAM and a 1.5GB pagefile, the pagefile utilization has never gone past 45 MB! I guarantee that if you try this on your machine, your pagefile utilization will be much, much, much smaller than you would expect.

Anyway, this is only a very brief description and I left a lot of the details out. Suffice it to say, Windows memory management is a really complicated topic to explain.
 
Sorry fallguy, but loadtimes are drastically increased with CCC and .net running for ATI. I adore my X1900XT, and find the drivers very stable (although don't like having to manually reload display driver through Rivatuner or Tray Tools to change AA/AF etc).

The boot time for my pc is however drastically lengthened due to .net and CCC. I keep my pc boot tweaked fast, with my HDDs kept defragged (Diskeeper) and system running optimum (Regsupreme, CCleaner etc), yet still have time to make a quick coffee while my pc boots.

As problems go its a small one, but still annoying. Why ATI can't keep the old CP as a backup alternative is beyond me.

Edit: Typos.
 
fallguy said:
If you're claiming CCC adds 45 seconds to your load time, then you are mistaken. Thats just pure lack of knowledge on your part.

Time to open Notepad.exe (added to Startup menu) with CCC installed, starting from the point I enter my password and click OK at the logon screen: 1:48

Time to open Notepad.exe (added to Startup menu) WITHOUT CCC installed, starting from the point I enter my password and click OK at the logon screen: 0:42

Time before SpeedFan 4.28 (in Startup menu) can begin to monitor readings from the onboard sensors with CCC: 1:22

Time before SpeedFan 4.28 can being to monitor readings from the onboard sensors without CCC: 0:23

Now, please, point out the "pure lack of knowledge" in the above...

I was just estimating 45 seconds before, it's even WORSE when I used a stopwatch. After your comment I took out the stopwatch and added Notepad.exe to Startup (it being one of the quickest loading .exe's I could find). And I'm on an X2 system, I suspect the load times would be even worse with a single core.

Enabling or disabling Crossfire had no impact, btw. Now, to further test it I suppose I could yank out my X1900xt secondary card, and see if the overhead is less with just the master card to deal with.
 
ManicOne said:
Sorry fallguy, but loadtimes are drastically increased with CCC and .net running for ATI. I adore my X1900XT, and find the drivers very stable (although don't like having to manually reload display driver through Rivatuner or Tray Tools to change AA/AF etc).

The poster didnt say anything about .Net, just that CCC added 45 seconds to his load time.

You dont have to do anything with Rivatuner or Tray Tools to change AA/AF. If you have CCC installed, just right click the icon in the task bar, and change it from there.

Croak said:
Now, please, point out the "pure lack of knowledge" in the above...

Because it is. If it took 45 seconds extra to load Windows, I would notice it, and so would many, many others. I have not seen one single review claim such a thing, and I have not noticed it myself. Next time you format, time your reboot. Then install .Net and CCC, and do it again. I guarantee you it wont add 45 seconds to your boot. Even if your claims were true, CCC wouldnt be the sole reason, .Net is not known for speeding up booting. There are many things that can slow a boot process, anti-virus, firewalls, etc. and claiming that CCC add 45 seconds is just rubish.
 
Croak said:
First off, I LOVE my X1900 Crossfire setup when I'm actually gaming.

But they've made the process of setting up a new overclocked rig a freakin' nightmare. Why? Not because they crash, not because they add a lot of heat to the loop, not because they give my 530w PSU a workout, and not because they're noisy.

It's because CCC adds FORTY FIVE SECONDS to my Windows XP load times...that really adds up when you're rebooting every few minutes to test a change, or crashing because you've pushed something too far, or just the restarts due to other software installs.

Seriously...45 seconds at XP load, AFTER getting logged on, with 50%-100% CPU load as CCC goes through its thing (and why does CCC have to hit my floppy drive?). This is with 2gb RAM, it's even worse with 1gb.

Hell, I've got an ancient HP 700mhz 256mb Celeron laptop that loads XP faster, and that's including loading all the HP power management apps, spyware and antivirus, HP update, wireless PCMIA card drivers, Bluetooth card drivers, etc.

A nVidia PCI-e 6200, or PCI MX card adds nothing to the load time overhead, ten seconds after XP login I'm fully stable and the only activity is XP doing a little diddling with the pagefile. Or even a single X1900 using ATI's older CP instead of CCC.

Unfortunately, I have NO choice in the matter, I don't have the option of using the old leaner meaner ATI CP, or ATI Tray Tools, etc, because the only way Crossfire will work is with CCC running.

It's not a big deal once you've got your system dialed in and aren't rebooting on a regular basis, but it still speaks volumes as to how bloated CCC is.

I feel the EXACT same way. It takes me a full 45 seconds longer at least to boot with this card than it did with the 7900GT after logging in. My computer is just slow as hell and it is annoying as fuck.

fallguy said:
If you're claiming CCC adds 45 seconds to your load time, then you are mistaken. Thats just pure lack of knowledge on your part.

No, it isn't a mistake. My experience with it has been enfuriating. I made sure to remove all remnants of nVidia's video drivers, control panels, overclocking utilities in preparation for installing this card. It still sucks.

Croak said:
Time to open Notepad.exe (added to Startup menu) with CCC installed, starting from the point I enter my password and click OK at the logon screen: 1:48

Time to open Notepad.exe (added to Startup menu) WITHOUT CCC installed, starting from the point I enter my password and click OK at the logon screen: 0:42

Time before SpeedFan 4.28 (in Startup menu) can begin to monitor readings from the onboard sensors with CCC: 1:22

Time before SpeedFan 4.28 can being to monitor readings from the onboard sensors without CCC: 0:23

Now, please, point out the "pure lack of knowledge" in the above...

I was just estimating 45 seconds before, it's even WORSE when I used a stopwatch. After your comment I took out the stopwatch and added Notepad.exe to Startup (it being one of the quickest loading .exe's I could find). And I'm on an X2 system, I suspect the load times would be even worse with a single core.

Enabling or disabling Crossfire had no impact, btw. Now, to further test it I suppose I could yank out my X1900xt secondary card, and see if the overhead is less with just the master card to deal with.

Perhaps you and I have a problem we can work mutually to fix...
 
Back
Top