New benchmark programme -Does anybody have a P3/1000?

[barryward12345 0]

This new benchmark programme by Maurizio Losso is absolutely IDEAL for people to do comparisons of similar systems - and particularly to see what hardware improvements will achieve for their framerates without having to purchase the piece of hardware first . Of course, it can also be used for the "mine is bigger than yours" game - but I will not enter into that arena.SO - does anybody have a P3/1000 with 512 meg SDRAM ?--- BUT with a better video card than my meagre Geforce2Mx with 32 meg onboard RAM. I am prepared to update my video card - but would like to see just exactly what I could expect in regard to improved framerates by doing so. I know that there is a commonly held view that updating a video card will not improve framerates -- but Maurizios' programme will prove or otherwise the relative validity of that.Thanks Barry

[barryward12345 0]

bump

[gosta 0]

Hi Barry,I've got an Athlon 1GHz, 512MB PC2700 RAM (running in sync with the CPU, i.e. 200MHz FSB) and a Radeon 9700 Pro. Will that do?Cheers,Gosta.http://hifi.avsim.net/activesky/images/wxrebeta.jpg

[barryward12345 0]

Thanks for the reply Gosta - but I think your processor/memory combination is probably faster than my P3/1000 133Mhx SDRAM system. So , I could not really compare my results with yours. Thanks for the offer though :)But anyway, I expect that you would have added the Radeon vid card sometime after the computer was built. If so, what vid card did you have before that and did you notice any improvement if/when you upgradedBarry

[Carabou 0]

Hi Barry,As of 2 days a go I had the same system as you described (PIII, 1 gig, 512 sdram, GeForce 2 mx400 32 mb). Yesterday, I installed an Asylum GeForce 4 Ti4200. Wow, what an improvement in fluidity and graphics. Not sure yet about fps improvement but there is some. What I noticed most is that I don't run into the "blurries" when flying slower aircraft like the Fleet 80 or the Cessna's. That was my big concern with my old card. Also, I can now truly enjoy some of the intense graphic scenery at some add on airports, which were barely useable before. The best example is Italy's Malpensa (2002) airport(www.isdproject.com) which is one of the best freeware airports I've seen. With the new card it is FANTASTIC especially at dawn, dusk or night!In any case, I also have the benchmark program and will run it in a few days. I'll let you know what I find. Howard

[John_Cillis 2,406]

I think almost certainly, the blurry improvement is due to the mem on your new card. 32 megs on your old card was a tad too low, I think. I went from worse--a 16 meg Voodoo. It was like coming out of the fog into a crystal clear, southwest morning....Oddly, owners who went from 64meg cards to 128meg cards haven't noted a huge improvement in blurries, although many also noted that the blurries they had were minor with both cards. But it makes me wonder if 2k2 doesn't recognize video memory beyond a certain point--64 megs. My upgrade experience makes me believe that the blurries had a huge relation to video memory. My 16 meg Voodoo that I began 2k2 with just wouldn't keep the sharpest textures, unless I held the visibility to 10 miles or less...

[gosta 0]

Hi Barry,yes, I upgraded one part at a time. First came the 9700 (a friend of mine had a deal that was too good to miss...), upgrading from a Radeon 7200 64MB, but running on an Asus A7S-LE mobo with 256MB of PC133 RAM. The improvement wasn't that much by way of actual fps numbers (maybe 4-5, from being around 18-20 to 20-25 average), but significant in terms of smoothness. Then I upgraded the motherboard and RAM to on Asus A7N8X Deluxe and 512MB of PC2700 RAM, and now I saw quite a noticable rise in fps in cruise (nearly double), but not that much on approach, although I'm not worried about handflying the RFP 747-200 into Simflyers' LAX at dusk, but then I'm only running AI at 30% (all PAI airlines installed, no GA traffic), scenery complexity at very dense and autogen at normal - the same settings I had before. However, this should change in a few weeks, when I'll get a new CPU :) (My wife insists I wait until my birthday...).So, from my experience, you should notice an improvement with a high-end graphics card, but it may be difficult to measure in actual numbers. But I would only recommend it if you're thinking of upgrading the rest of your computer as well in the near future, otherwise you may be better off spending the same amount of money on upgrading all your components to a slightly higher level, rather than getting one state-of-the-art part and leaving the rest where it is. In short, a newer middle-of-the-road GeForce and a 1.7GHz CPU may bring you more enjoyment than a 9700 or GeForce FX on your current system.Cheers,Gosta.http://hifi.avsim.net/activesky/images/wxrebeta.jpg

[barryward12345 0]

>Hi Barry, >>yes, I upgraded one part at a time. First came the 9700 .The improvement wasn't >that much by way of actual fps numbers (maybe 4-5, from >being around 18-20 to 20-25 averageActually, an improvement of "only 4-5 " is just what I am looking for!! :) Once you get over 20 (IMO) , sim flight is quite acceptable to me . But the difference between 13 and 18 is enormous!!Barry

[Guest PaulL01]

>I think almost certainly, the blurry improvement is due to >the mem on your new card. 32 megs on your old card was a >tad too low, I think. I went from worse--a 16 meg Voodoo. >It was like coming out of the fog into a crystal clear, >southwest morning.... >>Oddly, owners who went from 64meg cards to 128meg cards >haven't noted a huge improvement in blurries, although many >also noted that the blurries they had were minor with both >cards. But it makes me wonder if 2k2 doesn't recognize >video memory beyond a certain point--64 megs. My upgrade >experience makes me believe that the blurries had a huge >relation to video memory. My 16 meg Voodoo that I began 2k2 >with just wouldn't keep the sharpest textures, unless I held >the visibility to 10 miles or less... Hi John,I think you are mistaken on the Video memory as to blurry issue...I used to load up all of the 350mb+ custom textures from my old Rochester photorealistic scenery just fine with my old TnT16 card in FS2k2, the only "blurry" issue that a newer GF2-32mbGF3-64mb/GF4-128mb etc card will resolve is enabled and better use of anistrophy.As for "dynamic blurries" that is strictly a CPU/memory subsystem issue. That aside the higher Video memory certainly enabled seemless switching between views and panning as well as enabled better AA and higher screen resolution but thats it.I as well as others (many more qualified then self) Gaming gurues and FS scenery designers have tested this on most any video card/buffer size made and what is stated has become excepted and is pretty much a dead issue.Paul

[John_Cillis 2,406]

"That aside the higher Video memory certainly enabled seemless switching between views and panning as well as enabled better AA and higher screen resolution but thats it."My experience with the Voodoo may have been unique... The Voodoo could not display textures at their highest quality, unless vis. was cut below 10 miles. The fact that it used the AGP slot, but didn't exploit AGP texturing, may have been part of it. It didn't have this problem in FS2000 though, which I found interesting. But FS2000 had the hesitation, and I suspect that may have been the "sacrifice" we were stuck with for "always sharp" textures.I'm aware scenery designers and "gurus" have tested this. I consider myself a scenery designer by virtue of my own tools, and a guru by virtue of my being a WAN and Desktop Support Admin. But, with technology changing by the hour, I can't know everything. That's what I tell my newer techs, who figure earning their MCSE and A+ is worth an extra $10-20K at eval time.... :) With the GeF 2, depending on A/C complexity, I seldom see dynamic blurries. But take an aircraft like Flight 1's 421, or the Dreamfleet Cardinal, and I've seen a greater chance of dynamic blurries. And the poly count doesn't seem to correspond. I loaded up 1024x1024 textures on my Microlight while I was testing it for release. I definitely saw a change for the worse vs. the 256x256 textures I deployed it with. The Microlight looked better, but the ground looked a bit worse.So, I guess what I'm saying is I disagree. I believe, at least to a limit, graphics card texture memory comes into play. Now I do concede that it may be less of a quantity issue, and more of a speed issue. But I've analyzed the texture footprint of some of my scenery, like my photoreal Phoenix scenery, and that accounts for a huge load of textures, far more than what I imagine Rochester would take up. It's been a while since I've posted it, BTW.... I'll try to post some screenshots in the next few days....-JohnEdit: I should note that the photoreal Phoenix scenery I reference, which some may remember seeing shots of, I never released. I used the sat. photos off of mapquest, and stitched them together over a period of several weeks, eventually hitting over 400 sq. miles of NW Phoenix and its suburbs. I wanted to release it (once I figured out how to get around the bandwidth issues) and I contacted mapquest's sat. source vendor (Globexplorer), since they owned the copyright. They didn't know FS from a hole in the ground, and refused permission for a freeware release--well, didn't exactly refuse it, but after going up the sales chain, I just got ignored. So I'm stuck with the scenery and the odd screenshot I post from time to time.... I'll add some to the forums, perhaps with FFX's SWA 737 for "theme"...

[gosta 0]

Hi Barry,that's fine, as long as you don't expect too much - you will still get stutters when flying into very dense areas etc., and you most likely won't be able to increase your settings without a degradation in performance (except maybe screen resolution and anti-aliasing). But as I said, if you're thinking of upgrading the other components even as far away as six months down the line, I'd say go for it - otherwise, I'd check how much 'across-the-board' improvement you can get for your money.Cheers,Gosta. http://hifi.avsim.net/activesky/images/wxrebeta.jpg

[Guest PaulL01]

>My experience with the Voodoo may have been unique... The >Voodoo could not display textures at their highest quality, >unless vis. was cut below 10 miles. Not sure if this is what you are refering to but the Vodoo was/is limitted to 256x256 texturing and the GF2/3/4 will do up to 1024x1024.>With the GF2, depending on A/C complexity, I seldom see >dynamic blurries. But take an aircraft like Flight 1's 421, >or the Dreamfleet Cardinal, and I've seen a greater chance >of dynamic blurries. And the poly count doesn't seem to >correspond. I loaded up 1024x1024 textures on my Microlight >while I was testing it for release. I definitely saw a >change for the worse vs. the 256x256 textures I deployed it >with. The Microlight looked better, but the ground looked a >bit worse. Right John,Using a GF2 things like color, position, lighting, texturing, and shading besides the other geometric calculations yet to be ordered are STILL some what processed by the cpu on a GF2, add to that an inefficient use of the Subsystems Bus... hence even more "dynamic" blurries compared to a GF2-GTS, GF3, GF4 etc...So, while using your GF2 your CPU is still having to do extra work even if your just changing the size of the textures to be displayed and for that reason alone you should look into at the least a GF2-GTS(PCI) or a GF3(AGP)as it will help take a load off your CPU. >So, I guess what I'm saying is I disagree. I believe, at >least to a limit, graphics card texture memory comes into >play. Now I do concede that it may be less of a quantity >issue, and more of a speed issue. Very true hence DDR etc. But I've analyzed the >texture footprint of some of my scenery, like my photoreal >Phoenix scenery, and that accounts for a huge load of >textures, far more than what I imagine Rochester would take >up. >Edit: I should note that the photoreal Phoenix scenery I >reference, which some may remember seeing shots of, I never >released. I used the sat. photos off of mapquest, and >stitched them together over a period of several weeks, >eventually hitting over 400 sq. miles of NW Phoenix and its >suburbs. I remeber your very nice Phoenix scenery, Its full color texturing is rare.You may want to revisit the copy right issue again as I belive your manipulation and modifications would allow you certain rights for your work. Just a little insite into the Rochester project..Rochester for FS98 started out at over 8000sq miles and was/is based on 9x9 high res full color transparency sat image of the Fingerlakes region of upstate NY that was obtained at very high cost in the year 1999.In addition many full color ultra high res aerial images where also contracted out to a localy based firm that does work for the FAA.All of this data was professionaly drum scanned (wet) and burned on to CDs before my work could unfold.Rochester for FS2k2 was paired down to just over 1000sq miles and included two 30sq mile areas of actual 1meter textures using many 1024x1024 textures as well. In comparison to most any other scenery the dense Rochester photoreal stressed memory and the graphics buffer considerably. Night texturing as well as the other three seasons is another story... :-rollhttp://www.frontiernet.net/~pleatzaw/images/ecto3.jpg Take care John.Paul

[John_Cillis 2,406]

Love the Rochester shots, and thanks for remembering the Phoenix scenery. Didn't realize you caught so much of the surrounding area in your efforts--boy do I stand corrected! How the heck did you have time & patience to deal with the original 8000 sq. miles? And will you ever release it?I will look at the copyright issue again. Your mention of the manipulation and modification does trigger memory of a few posts on the issue.... Thanks.... I really would like to release the Phoenix photoreal, although I want to do some landclass touchup as well before I dare upload it... I think I managed to get it to around 20 meg zipped, but unzipped it comes in around 150 megs, I somewhat recall...Been a while since I played with photoreal, since I've been so busy with LCA. I'm currently trying to find some way to link to FS state data without going through FSUIPC.... That way, if Peter goes payware with his '04 release, it won't have a negative impact on my users....

[Carabou 0]

Just ran FS benchmark (twice) and here's the result:Avg. 11.93max peak 32.28max peak peak 30.20min peak 2.08My scores are slightly lower than yours - wonder why?My computer is made by Dell, who made yours?Anyway, I'm sorry I didn't run the test with my old video card to see the difference.Regardless of the benchmark test results, I'm very happy with the performance of my computer with FS2002 and the new Ti4200 card. I set my fps at 15 and have fluid motion, lack of blurries (most of the time) as well as extremely crisp video display.I agree with you that I can't justify replacing my computer at this time.Hope the above helps. Let me know if you want more information.Howard

[barryward12345 0]

>Just ran FS benchmark (twice) and here's the result: >>Avg. 11.93 >max peak 32.28 >max peak peak 30.20 >min peak 2.08 >>My scores are slightly lower than yours - wonder why? >My computer is made by Dell, who made yours? I put mine together myself about 18 months ago. I used an ASUS motherboard.>Anyway, I'm sorry I didn't run the test with my old video >card to see the difference. >>Regardless of the benchmark test results, I'm very happy >with the performance of my computer with FS2002 and the new >Ti4200 card. I set my fps at 15 and have fluid motion, >lack of blurries (most of the time) as well as extremely >crisp video display. Yes - I am reasonably happy with mine as well . It is only the very busy city areas like London that the computer doesn't like. I lock mine at 20 - and also don't get any blurries that others sometimes report. I may buy myself a better vid card sometime as the Geforce2 Mx was basically the bottom of the line at the time I built my machine - whereas the P3/1000 was top of the line at that time. So it was a bit of a mismatch right from the start.>>I agree with you that I can't justify replacing my computer >at this time. Agreed!!Barry