Jump to content
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble
Strawberry Orange Banana Lime Leaf Slate Sky Blueberry Grape Watermelon Chocolate Marble

MSFN is made available via donations, subscriptions and advertising revenue. The use of ad-blocking software hurts the site. Please disable ad-blocking software or set an exception for MSFN. Alternatively, register and become a site sponsor/subscriber and ads will be disabled automatically. 


Feamane

Member
  • Content Count

    45
  • Donations

    $10.00 
  • Joined

  • Last visited

  • Days Won

    1

Feamane last won the day on June 25

Feamane had the most liked content!

Community Reputation

3 Neutral

About Feamane

Profile Information

  • OS
    98SE
  • Country

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. I played these games a few years back LAN-party style (four players) on Win98 machines with minimum Athlon 64 2200MHz, 512MB RAM, and GeForce 6600. Game ran OK on these platforms, but I think you will have a lot of trouble with magic battles with slower machines. The end of HoU is buggy, not sure they ever fixed it--the game got so messed up at the end that I don't think we were ever able to fight the final battle. The Baldur's Gate series with lots of new community created mods is still fun to re-play on Win98! But even then my top end Win98 machines with P4-3400MHz, 1GB, GeForce 6800 can get an annoying amount of lag in big magic battles and/or with improved AI scripts running. That's why I'm trying to upgrade my Win98 boxes with Pentium Dual-Core E6800, PC2-5300 RAM, SATA SSDs, and GeForce 7800 cards. So far all is helping a little bit except I can't get any more performance out of the new GeForce cards than I did the old ones. Best regards, DJ
  2. I'd be super happy if I could get that score on this rig! I've been trying some things, all to no avail. I upgraded the memory from 512MB PC3200 DDR400 to 1GB PC2-5300 DDR2-667 and it improved the 3DMark 2001 score from 12829 to 14217, while I was getting 14338 with the eVGA nVIDIA GeForce 6800 and the 512MB of RAM. I tried changing the AGP aperture to various settings with no apparent affect. I also played with overclocking the card to Core 437 and Memory 703, but the difference in the benchmark score was negligible--so I don't think that's worth pursuing further. I tried installing an ASUS nVIDIA GeForce 7600GS 256MB, but neither the MDG nor the Zak 82.69 drivers will work with it, can't even finish booting windows once the drivers are installed. Also tried downgrading to the eVGA nVIDIA GeForce 6800 128MB card and I got a small bump in the score to 15082. This is with the official 81.98 drivers. So the memory and CPU upgrades allowed the 6800 to score about 300 points higher that it was before, yet I can't get the 7800 to perform even up to the level of the 6800. So it would seem that while the MDG 82.69 drivers allow the 7800GS card to work in this setup, they are not allowing the full potential of the card under win98. Are you installing any of the unofficial updates or anything like that which might be enabling you to get the most out of your card? Anybody else got any ideas? Thanks, DJ
  3. Good questions. For the CPU comparison: GeekBench 3 (Single core) Data courtesy Primate Labs Pentium E5300 1,358 Pentium 4 550 908 So yes, in single core the E5300 should be faster than the 550. I have not tried the 6800 with the new motherboard and CPU yet, I will have to try it. Good suggestion. I'm very curious why your nVidia control panel shows OS AGP Fast Write = Yes and mine doesn't. Do you use one of the unofficial services packs? Thanks, DJ
  4. Just to clarify the only differences between the old build and the new one. OLD ASRock 775V88 (VIA PT880) Via 4.56 driver set Pentium 4 550 3400MHz LGA775 512MB RAM SATA HDD 66MB/s avg. throughput speed according to DiskSpeed32 eVGA nVIDIA GeForce 6800 128MB nVIDIA 81.98 DirectX 9.0c 3DMark 2001 score 14338 NEW ASRock 4CoreDual-SATA2 R2.0 (VIA PT880 Pro/Ultra) Via 4.56 driver set (also tried the Via Pro5.15A AGP driver) Pentium Dual-Core E5300 2600MHz (should be about 17% faster according to UserBenchmark) 512MB RAM SATA SSD (with RLoew's ATA0104A) 99MB/s avg. throughput speed according to DiskSpeed32 BFG nVIDIA GeForce 7800GS OC 256MB Maximus-Decim's nVIDIA 82.69 Version 1.03 DirectX 9.0c 3DMark 2001 score 12829 Thanks, DJ
  5. It is good to know what kind of score I should be getting. Interesting that you show Fast Write as "YES" for the OS and I don't, what part of Win98 controls that? The one difference is I have 512MB instead of 2GB RAM, could that be the only reason for the big performance difference? I tried different AGP Aperture sizes but I receive the same test result so that doesn't seem to make any difference. I have 1GB of RAM on order, so waiting for that to arrive to try that. Any other ideas? Thanks, DJ
  6. Hi again, I am trying to build a new Win98SE game machine and have run into another problem. I have spent all day reading here and search the web, but have found no answers. Basic info: ASRock 4CoreDual-SATA2 R2.0 (VIA PT880 Pro/Ultra) with 512MB RAM BFG nVIDIA GeForce 7800GS OC 256MB DirectX 9.0c Thanks to the help here I have the SATA running with an SSD and am getting 99MB/s Avg speed according to DiskSpeed32, which is better than I've ever gotten before on any of my Win98 boxes. I also have the BFG nVIDIA GeForce 7800GS OC 256MB working using Maximus-Decim's nVIDIA 82.69 Version 1.03 drivers which I could never get to work before, so there is light at the end of the tunnel! I had tried the BFG NVIDIADisplayWin9x(82_16) and ZakMcKracken 82_69 Driver Edition 0.0.2.1 but neither of those worked for me. Now, I just have one problem remaining. When I run 3DMark 2001 I get a score of 12829 with the new setup, while I was getting 14338 before with the old eVGA nVIDIA GeForce 6800 128MB and SATA HDD (which got only 66MB/s on DiskSpeed32). "Vertical Sync" is turned OFF. Very disappointing to get a worse score! Trying to troubleshoot I looked at the AGP Settings tab in the NVIDIA Display Control Panel and see the following information: Capability GPU Chipset OS Optimal AGP version 3.0 3.5 3.0 3.0 Max AGP rate 8x 8x 2x 8x Fast writes Yes Yes No Yes So, is this information accurate? When it says "OS" does it mean the Via hardware driver for the AGP? Or is it referring to some other parts of Win98? Does anybody else get this same information from the NVIDIA Display Control Panel? I have been using the Via 4.56 driver set for the motherboard, and everything seems to be working. I also tried installing the Via Pro5.15A AGP driver as a test, but no difference either in the NVIDIA Display Control Panel or 3DMark 2001. So, is my problem that Win98 is not compatible with AGP 8x and Fast Writes? Do I need a different Via AGP driver? Or is there some other parts of Win98 that need to be upgraded? Thanks, DJ
  7. I tried both RLoew's PTCHSATA and Via SATA v6.10c. The difference was very similar according to DiskSpeed32. The Via drivers showed about 1MB/s faster Min and Max throughput, but the RLoew gave a 1MB/s faster overall average based on all file sizes.
  8. I've been experimenting with when to install it and when to change the system.ini settings myself. I've not arrived at any solid conclusions so far. But currently I'm installing the patch as soon as I can after the OS install and before the hardware drivers. Seems to be working OK, but I haven't even gotten past getting all the hardware drivers working correctly to think about applying the unofficial service pack yet, which may mess everything up.
  9. OK, I will have to do some testing to try both. But first I'm going to have to figure out why DiskSpeed32 keeps crashing on the new build, but that's a story for a different thread. Thanks! DJ
  10. Ah, OK. I tried the scandskw.exe from BHDD31.ZIP that I saw mentioned in some posts here, but same result. Now I know. I will look into Usher's method. I also have discovered that DiskSpeed32 3.0.0.5 is crashing on this new build which it doesn't on any of my other 1GB Win98 machines, which is why I was running the scandisk thinking it was maybe a disk problem. I'll have to run the scandisk from DOS and then try the DiskSpeed32 again. I need some way to measure my throughput so I can try some different settings and see if I can maximize what I have. I've been reading many posts, and have been using search as well. Of course sometimes the problem with search is you don't know the exact correct term to search for until you find the information you are looking for! But you are right, sometimes it is an iterative process since the first time through it can be like trying to take a sip from a fire hose. Thanks! DJ
  11. But xRayeR's patch for IO.SYS with a renamed Himemx is still needed for Safe Mode? This still leaves the Himem.sys to be used by Normal Mode, correct? In my current testing Safe Mode would hang until I did xRayeR's patch. But now when I try to do a Scandskw.exe on a 80GB HDD in Normal Mode it never is able to finish, it keeps hanging. Usually no message, but once it said that Windows was out of memory. This is a new install with 1GB RAM with Patchmem /M and the config.sys and system.ini changes mentioned in this thread. Ideas? Thanks, DJ
  12. Awesome! I was spinning my wheels reading all the different posts, it gets a little confusing going back into all the old threads--I now realize that while it was good I was taking notes as reading all the old threads, my mistake was not including dates in my notes.
  13. OK, thank you for that answer! Now that I have RLoew's SATA patch working I can forget about it and move forward with the project.
  14. If it is recommended to remove the MaxPhysPage, MinFileCache, and MaxFileCache when using PATCHMEM /M, what is the recommendation with this setup as far as using smartdrv.exe, xhdd.sys, and/or BUFFERS= setting in config.sys? There is a great deal of different ideas about using these scattered about in different threads here but I'm having trouble figuring out if there is any consensus on what the latest/greatest configuration would be when using PATCHMEM /M. What is the latest thought about using himem vs. himemx with this kind of set up? Thanks, DJ
×
×
  • Create New...