Jump to content
Sign in to follow this  
Noel

Opinions/facts on how HDD performance affects FSX performance

Recommended Posts

Guest Nick_N
Yes, I see that. Even so Nick, you make this comment in your article: "Velociraptors in RAID0 (256K STRIPE) on the right card are very close to SCSI performance"I have a 15K.5 SCSI right now already on a U320 controller (it's running 180 in single channel mode I think). The comment above suggests my SCSI should be ample for FSX, or am I misinterpreting something? I had been running FSX in XP on that drive, however I haven't moved my Vista 64 over to FSX on that SCSI, and was wondering if it would be worth it. Sounds like it may be, though again, it's isn't too bad on my sluggo SATA II drive.Thanks in advance . . .Noel
Hi NoelI am really pressed for time so I am going to make this fast, simple and without any techno babble to razzle and dazzle with flamboyant technical statements that suggest I may have a PhD in advanced toenail clipping :( 1. I did not say 2 Vraps in RAID are equal to SCSI in FSX.. I said in performance which is relevant to storage performance in a computer, not FSX. 2. I also said "2 Velociraptors in RAID0 (or SCSI/SAS) on the right card are overkill for todays processors and memory subsystems in conjunction with the video adapter in FSX and will deliver very large amounts of data faster than the system can process/render it"3. Yes I said 2 Vraps in RAID are close to SCSI which is true. Look at the chart and that is easy to figure out. The access time will drop slightly bringing it to within SCSI spec and the IO/sec will increase by about 35% or slightly better which means the 2 Vraps in RAID will then be just about equal to SCSI in a multi user IO environment4. There is a defined benefit to a larger platter over perf spec. That benefit drops significantly as the drive fills. That benefit will not ever replace or overcome the perf spec and ability however it does play into the equation. The loss to compare ratio is significantly increased by a very small platter storage device compared to the larger platter storage device. That is why I always specify the larger platter Vrap over the smaller one available and I would never consider a 30-70GB SCSI drive over a Vrap. The combination of the Vrap 300GB platter, the random access spec and the sustained IO perf ability place the drive at the top of the food chain when it comes to FSX and no single SAS or SCSI drive will beat it.. period.5. RAID is as useless as boobs on a bull to FSX if the controller card does not incorporate true hardware RAID with an on-card DDR memory cache, the card runs off a PCI and not a PCIe or PCIx slot, and, the ability to set a 256K STRIPE or BLOCK when the array is created. If those factors are in order then 2 modern 32mb cache drives in RAID on the right controller card will perform much better than any single drive other than the Vrap however the design of the single Vrap eliminated the need for those 2 32mb cache drives in RAID on a card and the cost of a single Vrap is cheaper than those drives + a controller card or any SCSI/SAS setup. Therefore it's a no-brainerIn terms of cost, the best storage performance for FSX is a single WD 300GB VelociraptorNo contestPlace it on the right controller card even as a single drive and its 'open season' on storage performance in compare. The card is not 'a requirement' as the Vrap will still do very well on motherboard ports but I would not run my single Vraps any other waySingle SCSI drives on a controller are essentially EQUAL (SCSI slightly better) to a SATA drive in IO performance... look at SINGLE user IO chart for SCSI and the 1TB drive Noel. They do sport better access and rotation however with respect to internal design and IO perf they are designed/engineered for multi user corporate environments, not gaming. Add in the much smaller platter size for the SCSI drive in a compare and the SCSI access gains begin to get lost in gaming between the internal design and basic geometry. Those small platter SCSI drives are very old news which is why the larger platter modern single drive is nearly equal in perf.. probably does beat them if the SCSI drive is nearly full.. any drive above 65% full and perf start dropping like a rock. The benchmark you are looking at is LINEAR read values.. again, useless to what FSX is which is a random read application not linear or sequential read.. Random read is not random access speed. Proper random read testing with IO values to establish ture application performance in relation to the storage solution requires IO Meter to evaluate and synthetic benchmark results are generic.Linear/sequential read tests such as Everest and HD Tach are for judging storage perf in systems delivering large single multimedia such as A/V editing, etc. That's all I have time for

Share this post


Link to post
Share on other sites
Guest Nick_N
Sam,Did your test include flying over a rural area into a very dense urban area with a big add on airport? I guess this change will result in a lot of disk access reading a lot of scenery and texture files. No difference at all in such a scenario?What about the ohter hardware when testing? What CPU, mobo, RAM, graphics card did you use? OC:ed?
The 'hit-n-run' expert doesn't get it. If he did, he would have nailed the fact that the SCSI drives in question here are very small platter and older design being compared to a much larger platter modern SATA, and, he would have known they are better suited for multi user IO. Instead what he did was read what Noel posted and took the opportunity to take a pot shot at me using Noels benchmark result as a means of backing up his lame test and claims. Goes hand in hand with his GTX 260 test months back where he said it did not perform any better than a 8800GT.. remember that one? Problem is, he installed the first revision defective neutered version that Nvidia replaced on the market a few months later and he also put that card in a system that was already overwhelmed so there was nothing that card WOULD have done for him. Yet we had a wonderful 5 paragraph technical dysentery (yes.. that’s dysentery, not dissertation ) and a slew of follow up muck in thread after thread on why that card and all others above the 8800GT are useless to FSX.. If Noel was on a 8800GT he would still be having sound crackle issues. Forget the core advancements or the memory buss speed, the amount of the memory on the card allows him to set a large bufferpool level effectively dropping the latency to the buss through resource request at the driver level and fixing the sound crackle issue he was plagued with when pushing autogen sliders or running large autogen addons such as FTX. I happen to know for a fact that on higher sliders FSX well exceeds the 20% reserve PCIe bandwidth for autogen which is what causes all sorts of problems and is what the larger/faster video memory cards address as well as cloud rendering.Phil Taylor did not post what he did about higher speed/amount VM cards to BS the community Unfortunately it simply proves further he has no real tech ability past reading the internet hardware sites and a decent grasp of the language.. AKA: A Marketing Expert

Share this post


Link to post
Share on other sites

Nick, do you have some recommendation on the SATA Controller Card?As I mentioned in another thread, I was thinking about buying Vrap 300GB. Now thinking about the separate card if its not too expensive. How much better performance could I expect out of it?Could you also give your opinion to this thread?http://forums1.avsim.net/index.php?showtopic=246195Much appreciated!

Share this post


Link to post
Share on other sites
Guest UlfB
The 'hit-n-run' expert doesn't get it. If he did, he would have nailed the fact that the SCSI drives in question here are very small platter and older design being compared to a much larger platter modern SATA, and, he would have known they are better suited for multi user IO. Instead what he did was read what Noel posted and took the opportunity to take a pot shot at me using Noels benchmark result as a means of backing up his lame test and claims.
Problem is he might fool people here on avsim to make the wrong decisions on what to buy. It's a shame.

Share this post


Link to post
Share on other sites
Guest Nick_N
Nick, do you have some recommendation on the SATA Controller Card?As I mentioned in another thread, I was thinking about buying Vrap 300GB. Now thinking about the separate card if its not too expensive. How much better performance could I expect out of it?Could you also give your opinion to this thread?http://forums1.avsim.net/index.php?showtopic=246195Much appreciated!
If you are a simmer who wishes to have the best storage solution consider the drive and/or the controller card an investment to supplying the system what it needs now and in the future. As I posted earlier in this thread performance in MSFS (all versions) is a combination of components working in concert to produce the result. There is no single component that fixes everything. Those who look for magic bullets in one part or another always find themselves let down however those who carefully consider their component choices and realize they are working to a primary goal of a complete hi-perf system reap the benefit of their choices. I do not know what it is you expect. As people have posted the storage solution is not going to make the frame rate at busy hubs increase. Well, that is true when you consider the storage system alone however when the right optimize routine is applied in so much as the defrag process, and please do not let anyone tell you that makes no difference, then you can see better frame rates and frame transitions globally. My suggestions around the use of O&O Defrag and setting up Windows to allow O&O to do its job are key to that type of perf increase. http://www.simforums.com/forums/forum_post...D=168400#168400As for the card and its settings.... http://www.simforums.com/forums/forum_posts.asp?TID=29330What it will do is eliminate file access/read stutters, and, the right solutions remove overhead from the CPU. The argument that

Share this post


Link to post
Share on other sites
Guest Nick_N
The controller card is not a requirement however it too presents a benefits. Those who have moved to such solutions will tell you they would never go back to motherboard SATA
Given the less than expected Random Access Read stats I was getting I bit the bullet and redid the disk at 64K NTFS format from scratch. Ditched UltimateDefrag and went back to O&O.Total bliss. She's slammin' fast. (And even Frame Rates are up) .So I just had some weird file corruption that was slowing things down.Thanks so much Nick, for the 3ware 9650SE push.

Share this post


Link to post
Share on other sites
Guest D17S

The real trick is to not believe anyone. Use your own eyes and be your own judge . . . and try to forbear the typically profligate response. There seems to be some other circumstance of which we are not aware. That inimical nature is difficult to rationalize. In any case, the results will be Extremely close. There is no FS magic available, yet. A Q6600@3.6/4G-ram@800/9800GT/7200RPM-HD will provide FS performance will be virtually indistinguishable from an i7/965@stock/6G-ram@1600/GTX295/Raptor-raid

Share this post


Link to post
Share on other sites
Hi NoelI am really pressed for time so I am going to make this fast, simple and without any techno babble to razzle and dazzle with flamboyant technical statements
Nick, if that was supposed to be without flamboyant statements I wonder what the full court press would look like?!OK, I'm going to get real concrete now: let's take the Cheetah 15K drive at 1/2 full, versus the Velociraptor at barely full. Exactly, HOW will we see evidence of the performance (and you know how I define performance) benefit of one drive over the other? This is the only meaningful question UNLESS the only consideration is in "having the best" no matter what the cost. When I decide where to go with a purchasing decision, I try to get at least a crude sense of what the cost:benefit ratio looks like, even though I obviously don't always get enough information to make the best decisions. If we look at one performance measure and called "smoothness" or "freedom from stuttering", we could actually quantify that as #'s of subjective perceptions of stuttering per minute, and compare one against the other. Or, blurry textures as a percentage of the entire display in a controlled configuration. These things could be quantified, and one could say conclusively, "in this study under these conditions the V-rapt has 40% less stutters per minute . . . yada yada.OK, without asking for quantification, give me your subjective assessment of the impact of a Vrapt drive over my old news Cheetah. I think it's a very valid question if you have an opinion Nick and everyone else.Thanks bunches . . .

Noel

System:  9900K@5.0gHz@1.23v all cores, MSI MPG Z390M GAMING EDGE AC, Noctua NH-D15S w/ steady supply of 40-60F ambient air intake, Corsair Vengeance 32Gb LPX 3200mHz DDR4, Sabrent NVMe 2Tb x 2, RTX 4090 FE, Corsair RM 850W PSU, Win10 Pro, LG Ultra Curved Gsync Ultimate 3440x1440, TCA Boeing Edition Yoke & TQ, Cessna Trim Wheel, RTSS Framerate Limiter w/ Edge Sync for near zero Frametime Variance achieving ultra-fluid animation at lower frame rates.

Aircraft used in A Pilot's Life V2:  PMDG 738, Aerosoft CRJ700, FBW A320NX, WT 787X

 

Share this post


Link to post
Share on other sites
Guest simjunkie
The real trick is to not believe anyone. Use your own eyes and be your own judge . . . and try to forbear the typically profligate response. There seems to be some other circumstance of which we are not aware. That inimical nature is difficult to rationalize. In any case, the results will be Extremely close. There is no FS magic available, yet. A Q6600@3.6/4G-ram@800/9800GT/7200RPM-HD will provide FS performance will be virtually indistinguishable from an i7/965@stock/6G-ram@1600/GTX295/Raptor-raid
God you are an arrogant, disrespectful chump Sam. :( You need to learn how to respect people who have the knowledge you are too dense to learn and many, many, many, MANY MORE years of experience than you. How can you be so blind???? Really! The more you babble on and on the more people here see what a quack you are. I would be far more polite to you except for the fact that you litterally spit in Nick's face. You also lead people astray with WRONG ADVICE! How you lasted so long here I'll never figure out. :( -jk

Share this post


Link to post
Share on other sites
I ran some FS tests with HD transfer rates of ~ 250MB/s then with at 100MB/s. There was no perceivable ingame result and load times were only marginally affected. Overall, my system "felt" more responsive with the 250MB/s transfer configuration. but this was sensed only because one tends to develop a very tactile sense about their own system. Again, the FS change was indecipherable. These test indicate that HD transfer rates at ~ 100MB/s are more than FS needs for any aspect ingame play. Faster transfer rates might help game/flight load times, but by only a few seconds, at best.
Thanks Sam for giving me your impression about what the perceivable ingame result and load times in your particular situation. I was hoping to get something actually a little more subjective from people with the highest end machines, but so far most of the replies are about what works best. No offense intended to Sarg and Bert and Nick, I'm actually most interested in the differences between my particular 15K SCSI or the SATA II for that matter, and the Vrapt that is being promoted in terms of end-user subjective experience, since that is what matters to me. I don't know what to make of that, but thank you again for posting BOTH your opinions or experiences as well as some theoretical comments.

Noel

System:  9900K@5.0gHz@1.23v all cores, MSI MPG Z390M GAMING EDGE AC, Noctua NH-D15S w/ steady supply of 40-60F ambient air intake, Corsair Vengeance 32Gb LPX 3200mHz DDR4, Sabrent NVMe 2Tb x 2, RTX 4090 FE, Corsair RM 850W PSU, Win10 Pro, LG Ultra Curved Gsync Ultimate 3440x1440, TCA Boeing Edition Yoke & TQ, Cessna Trim Wheel, RTSS Framerate Limiter w/ Edge Sync for near zero Frametime Variance achieving ultra-fluid animation at lower frame rates.

Aircraft used in A Pilot's Life V2:  PMDG 738, Aerosoft CRJ700, FBW A320NX, WT 787X

 

Share this post


Link to post
Share on other sites
Guest D17S

That's the situation I'm seeing too. The change in FS performance improvement (on which these somewhat animated!) discussions proceed are so small that it becomes an entirely subjective assessment. This is not to say these performance gains do not exist, only that they are soooo minor that it becomes virtually impossible to even describe. For instance, that's why I like the acronym UFPS (unique frames per second). That describes a visual feel that (appears) to correspond to the frame counter. If I see 20FPS and the look is visually smooth, those 20 frame are Unique. Frames are not repeating ie, causing the "stutters." I don't have any problems with the default airplanes. All this scenery loading improvement chatter appears to me to be a solution in search of a problem. Once these potential improvement reduce to point that they are virtually invisible to a general-gamer, (ie, for me) these tweaks becomes more about 'fun with computers' rather than, 'Ahhh, I think that helped' . . . via multi-hundred dollar expenditures. Specs are helpful, but in the end, it's what a player sees on the screen. Hundreds-of-dollars spent also tends to invoke that placebo effect. Gotta watch out for that too. Unless I can see an actual, definitive difference, the cost/benefit quotient doesn't justify the expense. 'Ahhh, I think that helped' is not enough (Ahhh, for me, that is!)

Share this post


Link to post
Share on other sites
Guest simjunkie
That's the situation I'm seeing too. The change in FS performance improvement (on which these somewhat animated!) discussions proceed are so small that it becomes an entirely subjective assessment. This is not to say these performance gains do not exist, only that they are soooo minor that it becomes virtually impossible to even describe. For instance, that's why I like the acronym UFPS (unique frames per second). That describes a visual feel that (appears) to correspond to the frame counter. If I see 20FPS and the look is visually smooth, those 20 frame are Unique. Frames are not repeating ie, causing the "stutters." I don't have any problems with the default airplanes. All this scenery loading improvement chatter appears to me to be a solution in search of a problem. Once these potential improvement reduce to point that they are virtually invisible to a general-gamer, (ie, for me) these tweaks becomes more about 'fun with computers' rather than, 'Ahhh, I think that helped' . . . via multi-hundred dollar expenditures. Specs are helpful, but in the end, it's what a player sees on the screen. Hundreds-of-dollars spent also tends to invoke that placebo effect. Gotta watch out for that too. Unless I can see an actual, definitive difference, the cost/benefit quotient doesn't justify the expense. 'Ahhh, I think that helped' is not enough (Ahhh, for me, that is!)
HA! :( You're jealous. You don't have the chops to compete Sam and none of your arguments hold water. I haven't seen you ONCE present anything other than some stupid FPS argument. That and a bunch of jibberish.The decent and correct thing for you to profess would be that defined improvements are there to be had...AS PROVEN AND DEMONSTRATED BY THE ENGINEERS... but it requires the right HW setup and likely more expense. Instead you note cost alone and poo poo everything scientific like a jealous, stubborn fool. I'll ask again: What are your qualifications to make the statements that you do? What is your degree in? How many years of experience do you have in computer science? What contributions and consulting advice do you give to the industry?Answer these questions Sam before you go on trying to discredit Nick or anyone else with similar expertise!-jk

Share this post


Link to post
Share on other sites
That's the situation I'm seeing too. The change in FS performance improvement (on which these somewhat animated!) discussions proceed are so small that it becomes an entirely subjective assessment. This is not to say these performance gains do not exist, only that they are soooo minor that it becomes virtually impossible to even describe. For instance, that's why I like the acronym UFPS (unique frames per second). That describes a visual feel that (appears) to correspond to the frame counter. If I see 20FPS and the look is visually smooth, those 20 frame are Unique. Frames are not repeating ie, causing the "stutters." I don't have any problems with the default airplanes. All this scenery loading improvement chatter appears to me to be a solution in search of a problem. Once these potential improvement reduce to point that they are virtually invisible to a general-gamer, (ie, for me) these tweaks becomes more about 'fun with computers' rather than, 'Ahhh, I think that helped' . . . via multi-hundred dollar expenditures. Specs are helpful, but in the end, it's what a player sees on the screen. Hundreds-of-dollars spent also tends to invoke that placebo effect. Gotta watch out for that too. Unless I can see an actual, definitive difference, the cost/benefit quotient doesn't justify the expense. 'Ahhh, I think that helped' is not enough (Ahhh, for me, that is!)
Sam my impression is that with the machine I have now, with FPS_Limiter set at around 30 and FSX at 30, I am getting downright amazing IQ and performance. What's more I was able to make a full size FRAPS with ATC, music, Live ATC and all the add ons and it looks and runs totally smooth and awesome! I just did a recording a few minutes ago and it was amazing. I am quite impressed with my hardware now, esp with Vista 64 plus FPS_Limiter. Have you tried it Sam?--it's been a super find for me.
HA! :( You're jealous. You don't have the chops to compete Sam and none of your arguments hold water. I haven't seen you ONCE present anything other than some stupid FPS argument. That and a bunch of jibberish.The decent and correct thing for you to profess would be that defined improvements are there to be had...AS PROVEN AND DEMONSTRATED BY THE ENGINEERS... but it requires the right HW setup and likely more expense. Instead you note cost alone and poo poo everything scientific like a jealous, stubborn fool. I'll ask again: What are your qualifications to make the statements that you do? What is your degree in? How many years of experience do you have in computer science? What contributions and consulting advice do you give to the industry?Answer these questions Sam before you go on trying to discredit Nick or anyone else with similar expertise!-jk
Come on now SimJ he's entitled to his opinions and ideas and emphasis as much as the next guy. The fact remains: one can build a machine and get a fair amount of what FSX has to offer for hugely less money than many of us spend, and so the Cost:Performance goes up and up: you often have a huge premium to close in on perfection in FSX. I could afford to, am glad I did, but I'm glad one can make the sim useable and enjoyable at much less cost. Even the highest end machine can be stressed with FSX, so at some point, it's very relative, and Sam's point has always been: bang for buck matters--at least to some.

Noel

System:  9900K@5.0gHz@1.23v all cores, MSI MPG Z390M GAMING EDGE AC, Noctua NH-D15S w/ steady supply of 40-60F ambient air intake, Corsair Vengeance 32Gb LPX 3200mHz DDR4, Sabrent NVMe 2Tb x 2, RTX 4090 FE, Corsair RM 850W PSU, Win10 Pro, LG Ultra Curved Gsync Ultimate 3440x1440, TCA Boeing Edition Yoke & TQ, Cessna Trim Wheel, RTSS Framerate Limiter w/ Edge Sync for near zero Frametime Variance achieving ultra-fluid animation at lower frame rates.

Aircraft used in A Pilot's Life V2:  PMDG 738, Aerosoft CRJ700, FBW A320NX, WT 787X

 

Share this post


Link to post
Share on other sites
Guest simjunkie
Sam my impression is that with the machine I have now, with FPS_Limiter set at around 30 and FSX at 30, I am getting downright amazing IQ and performance. What's more I was able to make a full size FRAPS with ATC, music, Live ATC and all the add ons and it looks and runs totally smooth and awesome! I just did a recording a few minutes ago and it was amazing. I am quite impressed with my hardware now, esp with Vista 64 plus FPS_Limiter. Have you tried it Sam?--it's been a super find for me.Come on now SimJ he's entitled to his opinions and ideas and emphasis as much as the next guy. The fact remains: one can build a machine and get a fair amount of what FSX has to offer for hugely less money than many of us spend, and so the Cost:Performance goes up and up: you often have a huge premium to close in on perfection in FSX. I could afford to, am glad I did, but I'm glad one can make the sim useable and enjoyable at much less cost. Even the highest end machine can be stressed with FSX, so at some point, it's very relative, and Sam's point has always been: bang for buck matters--at least to some.
Noel,Thank you for being patient but I watched this go on for many months. I totally agree that one does not need the top of the line HW to get a good sim experience, and that extra polish comes at a substantial cost. It's not that Sam's opinion is that you can get decent performance with less. It's that he continuously states that the hardware and tuning is worthless and ineffective with absolutely no basis in fact. And in doing so he takes great pleasure in insulting the person who has helped more than anyone else in the FSX community. There is no acknowledgement that it is providing a benefit in spite of the fact that many have seen the improvements and more importantly...the actual professionals who know far more about it then he does have shown this to be true. And when challenged to back up his assumptions he has nothing to say until his next chance to show up in a different post somewhere else. If Sam would actually admit that SOMEONE here knows more than him about computer HW and FSX this wouldn't be an issue. If he is going to comment on what an i7 can or cannot do he should at least have one. I've had all the parts that I comment on, and I make every attempt to accurately represent advice given by accredited professionals, not just what I've experienced on my own. I don't just state that because I do or do not see anything, that advice is right or wrong. Sam doesn't just state his opinion here, he spits in the face of experts the way he writes his arguments. It's rude, it's misleading, it's disrespectful.-jk

Share this post


Link to post
Share on other sites

I ask you Sam for the last time to stop this "ping-pong" game with NickN if you want to remain a member of this forum.This is frankly boring and immatureI really hope you'll understand this warning.ThxSimjunkie, please don't make things worse everytime too.OK?


Best regards,
David Roch

AMD Ryzen 5950X //  Asus ROG CROSSHAIR VIII EXTREME //  32Gb Corsair Vengeance DDR4 4000 MHz CL17 //  EVGA GeForce RTX 3090 FTW3 ULTRA GAMING //  2x SSD 1Tb Corsair MP600 PCI-E4 NVM //  Corsair 1600W PSU & Samsung Odyssey Arc 55" curved monitor
Thrustmaster Controllers: TCA Yoke Pack Boeing Edition + TCA Captain Pack Airbus Edition + Pendular Rudder.

 

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  

  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...