FSX Dynamic Benchmark Run Results

[Guest D17S]

I've got that 9800GTX+ on the way. It was on sale. There are bios modding tools that let one set-and-forget. 800/2000Mhz are my target GPU/Memory clocks. See, I'm guilty too. I spent $50 on a 600 > 800 GPU clock increase. I'm giving up food for the next 2 days, but I might get thirsty! Maybe I'll just wander over to the neighbors. The art of K-mart shoppin' can have it's rewards.http://www.matanzascreek.com/

[Great Ozzie 101]

>although Noel and I may not see eye to eye on that cooling method he was using (window AC Unit which DOES present possibilities to why his motherboard failed... not conclusive but scientifically sound)Dang Nick

[Guest Nick_N]

last post for me in this thread Ozz ... I apologize to the forum for being quite off-key short and abrupt however I am fed up and in that I said too much the wrong way already. quite franky I have had enough of the smart-remark know-it-all 10lb word peanut galleryAbout AC systems and computers. Micro-condensation can form on the motherboard, inside sockets and around traces on the board and on cards from a process known as thermal inversion. The result of temp swings in an uncontrolled environmental siuatation like that can create all sorts of errata and strange problems, or catastrophic failures.Unless an engineer sits down and calculates the flow to ambient and internal temp changes with variable humidity factors moisture can in fact form -OR- the flips side, very high and dangerous amounts of static electricity if the air is too dry. A computer room does not use standard AC systems. Such systems come with special equipment and a flow design method to maintain humidity levels so sensitive SMD electronics and traces are not damaged or influenced by the cooling process. They can be called CRAC and AHU systems. They can have humidifiers/dehumidifiers and reheat coil systems which play in maintaining the correct humidity level which in turn controls static electricity. Temp/humidity/flow sensors maintain the correct inlet and outlet env factors based on changes during the use of the system(s) or their load status which is how the components are kept cool without the danger of any moisture buildup or static.As far as I am concerned, attaching the air output of a standard room AC unit directly to a computer is a no-no. I can see someone playing with something like that if they are willing to accept the risks of loss however I would never run something like that on any mission critical system or system I was not willing to lose.Using the evap tube as a phase change cooling unit is OK since that does not fill the tower with unchecked airflow however that process also requires a heavy amount of insulation on the tube and silicone at the CPU socket to prevent icing and condensation buildup from making contact with the components. you may review a simplistic explanation of the cooling process and caveats of standard AC systems in cooling sensitive computer electronics here:http://www.apcmedia.com/salestools/VAVR-5UDSLG_R2_EN.pdf

[Guest D17S]

Of course there's always the "proof's in the puddin" evidence to to consider. If it's been working for years, one might want to reconsider an obviously myopic assessment . . . and if the arguer ever resorts to an Ad Hominem approach to support any argument, an audience might wisely reconsider the veracity any additional presentations.

[Noel 4,563]

>As far as I am concerned, attaching the air output of a>standard room AC unit directly to a computer is a no-no. I can>see someone playing with something like that if they are>willing to accept the risks of loss however I would never run>something like that on any mission critical system or system I>was not willing to lose.Thank you for making this distinction. I have been blowing cool air over all the components of my systems now for a total of 15 PC years. That is 3 systems, running for 5 years each. Each one had a small thru the wall AC unit, about 8-9BTU's, and air is routed via aluminum flex hoses directly at the CPU from point blank range, and the other two 5" hoses are directed at ram, mainboard and video card. I have had one mainboard fail, and it was being overclocked considerably above its default rated FSB, and without a boost in NB voltage. The board has been plagued by failures, so in keeping with Occam's Razor, I'm not about to invoke microcondensation as the culprit when there are other plausible explanations. I had two drives fail in 15y. One video card. All the while, cool air is blowing away.I think APC's article is fine for mission critical environments. My system is not mission critical. I also have measured the temp of air that is coming out of those hoses: about 54-60 degrees by the time it gets to its target. The case and rest of the system DOES get quite cool. But I can tell that in this very low humidity environment, with these temperatures, blowing dehumidified air directly at a fairly LOW FLOW rate on components that are warm to the touch, the likelihood of micro or macrocondesation is about negative 80. Sorry, just ain't going to happen. I don't need a degree in air conditioning to deduce this. The net result is that I have enjoyed overclocked systems using stock HSF's, for 15 years, with no more troubles than the next overclocking vidiot, so it's quite easy to deliver a stream of air that is predicably cool & dry, not extreme. Oooo, AC, AC bad! Not!15y of emperical evidence trumps APC (the precision air conditioning vendor?)'s eloquent descriptor of the perfect, precision air conditioner for mission critical systems.

[Noel 4,563]

Sam did you see the recent comparison of video cards in FSX SP2?What the heck are you pick up at 9800GTX for? Same 256-bit, same GDDR3 memory, same memory amount? Why bother? When you look at the Tom's article the 8800GT or 8800GT 1Gb card looks very attractive at my monitor settings. I still want an ATI card that can handle FSX. QX9650 w/ Retail HSF|ASUS P5E3 Premium WiFi|4GB Muskin Ascent 7-6-6-18 1T DDR3-1600|EVGA 8800GT|Seagate SATA 2 x 2|Seagate Cheetah 15K.x|XP Pro SP2|Vista 64--maybe never to be installed

[Guest D17S]

Ya, you're right. I'm fasting now (vEryy happpilly). It was really the toy value of being able to get that 800Mhz. But I've O/C'd 8800s and thought I saw a difference in hi-Vload environments, but this was only only on the basis of a GPU clock increase.But with the default airplanes, All sliders are already maxed, sans water at 2XL/shadows/bloom and a one notch 'give-back' of AG. Traffic at 20/20/20/10/20/20. This ol' Q@3.6/DDR800/8800GT rig has no problem with the default product. I have no issue with clouds and the default airplanes. Don't even run the frame counter. It runs fine.The PMDG/LevelD airplanes are another story. Below 10K, all other settings can remain but I have to give back all but one notch of AG and traffic goes to 10/0/0/0/0/0/0/0 (more for ATC chatter than anything else).This 386bit memory interface story simply does not ring true in light of the non-subjective, reasoned evidence. The GTX2X0s should be showing incontrovertible improvements across the board because of their memory interface and increased memory size. They're not.

[Guest Nick_N]

> . . and if the arguer ever resorts to an Ad Hominem approach to support any argument, an audience might wisely reconsider the veracity any additional presentations.http://forums.avsim.net/dcboard.php?az=sho..._id=43245&page=Mel Brooks had a name for some of the recommendations and technical reasoning posted for those recommendations around hereStandup PhilosopherThis is an individual who places higher emphasis on their vocabulary than their ability to properly assess and conclude. They have technical ability but with a stronger knowledge of the terminology in definition within subject matter, perhaps assisted by a formal education on the subjects, which they rely on to impress and draw attention to their communications but their conclusions tend to follow the same pattern: They repeatedly demonstrate the individual can not access and coordinate the subject in an objective way or in a way which encompasses all the facts and leaves out facts from authoritative sources who are ---far-- more qualified to draw conclusions within the subject they are discussing than themselves. Therefore conclusions are drawn from their limited ability in observation and ego but in reality all they have done is convinced themselves and anyone else who may be impressed with the terminology and statements they use, the person knows what they are talking about and are an expert.AKA: Bullchit Artist our communications on this forum are concluded.

[Guest Nick_N]

> If bus speed is key to graphic card edge, why not use>Nvidia's 260 or 280 cards,(448 bit and 512 bit memory buses>respectively vs 384 bit for Ultra)? And both have more memory>that the 8800 Ultra (896 mb and 1 gb vs 768 mb)? As for>driver issues, I'm getting used to the black squares and>rectangles when I pan too fast with just about every version>of Nvidia drver software.http://forums.avsim.net/dcboard.php?az=sho..._id=43249&page=black squares (in part) indicate the system is working to catch up and they were influenced to appear by something that was introduced with the SP updates.. exactly what I can not say,.. however you are correct in your observation that getting rid of them is quite tough... under high load conditions in and around large hubs with the settings I use I still see them from time to time but I do wish to emphasis that when I compare their appearances on DDR2 and DDR3 systems and the better quad with better video card support (384bit) on the DDR3 system the time those black squares may appear on my screen are significantly reduced to what amounts to a quick flash now over a long pause before disappearing and showing the lights in the distance or on the aircraft which is the source of those black squares.. the light textures.Getting back to the card.. yes the 280 is a better choice than the Ultra in FSX with recent driver changes and improvements but it does come with a cost. If one has the money it would be the best choice. If not, the 384bit 768MB Ultra is the best bang for the buck in FSXBuss speed is not the only key but in FSX it holds a very large chunk of the perf because of how FSX handles memory and draw calls http://forums.avsim.net/dcboard.php?az=sho..._id=43245&page=Where many games may not show the perf increase under certain scenery situations with larger bandwidth memory buss, FSX certainly will.And for considering the 260 over the 280 keep in mind the 260 is a card made from a 200 series 'speed bin' tested and categorized component which was rejected for the 280 bin when it failed to pass the 280 QC performance tests stable at Nvidia.. its the bargain version and as such you will get what you pay for regardless of the memory bandwidth.if you are going to spend the money, do so and be truly satisfied for a long period of time. you must have a PCI 2.0 or above system to have all the benefits of the 200 series cards. The Nvidia selector assistance system on their website will not even show you the card if that standard is not selected as one of your systems attributes.

[Noel 4,563]

>if you are going to spend the money, do so and be truly>satisfied for a long period of time. Well, I'm trying. But I was really hoping for ATI as you know. I'm going to hang on a little longer in the hopes the 4870 or 4870x2 begins to shine better in FSX. If it could do what my current 8800GT can do right now, I'd grab it right away. Even an x2 I would consider.>you must have a PCI 2.0 or above system to have all the>benefits of the 200 series cards. The Nvidia selector>assistance system on their website will not even show you the>card if that standard is not selected as one of your systems>attributes.If driver releases don't improve the picture in FSX I will abandon the 4800 ATI's and consider the 280GTX. It sure has promise from a theoretical standpoint. You say it is now showing improvements in FSX, with recent driver release? I have a PCI 2.0 system.

[Guest djt01]

I'm going to hang on a little longer in the hopes the>4870 or 4870x2 begins to shine better in FSX. If it could do>what my current 8800GT can do right now, I'd grab it right>away. Even an x2 I would consider.How would a 4870X2 or even a 4870 be less capable in FSX or anything else for that matter, than an 8800GT?

[Guest Nick_N]

I know you want that ATi hardware image change pretty bad, but I dont think at this point in time you would benefit (in a huge way) from going from what you have to that card in FSX.. it may get better with driver updates but its really better suited for other games 'right now' than FSX in overall perf ability as compared to the Ultra or 280. I do think the x2 will do you better than the 8800GT but I also think it needs a bit more driver time before that will be worth the swap for an FSX only cardIf its for more than FSX, then the x2 may be a reasonable switch for you if that is what you would prefer to run.I do think the 280 is coming out of its shell with the latest 177.92 beta driver but if your primary game is FSX and you have a decent card I still think waiting a bit is a better choice. My suggestion to those who wish to upgrade to go for that card are:A: You are coming from a much slower card nowB: You intend to play much more than FSXC: Its an upgrade you intend to hold on to for a while to absorb the costD: You have the PCI 2.0 interface on your motherboardOtherwise I would wait a bit longer just to make sure what is being seen with FSX remains on trackTime also usually translates to a bit lower priceNow, if money does not matter and you play games other than FSX and you have the 2.0 slot, go for it as it will improve your overall gaming experience.the 280 has more driver dev time than the x2 right now so thats another reason I suggest waiting. The x2 was just released in dev terms of driver corrections and fixes

[Noel 4,563]

>I'm going to hang on a little longer in the hopes the>>4870 or 4870x2 begins to shine better in FSX. If it could>do>>what my current 8800GT can do right now, I'd grab it right>>away. Even an x2 I would consider.>>How would a 4870X2 or even a 4870 be less capable in FSX or>anything else for that matter, than an 8800GT? I was going by benchmarks, and a few odd postings I've read on this. Tom's for example puts the 4800's in FSX as pretty weak, QUITE weak. I guess you have one, and don't concur?QX9650 w/ Retail HSF|ASUS P5E3 Premium WiFi|4GB Muskin Ascent 7-6-6-18 1T DDR3-1600|EVGA 8800GT|Seagate SATA 2 x 2|Seagate Cheetah 15K.x|XP Pro SP2|Vista 64--maybe never to be installed

[Noel 4,563]

>the 280 has more driver dev time than the x2 right now so>thats another reason I suggest waiting. The x2 was just>released in dev terms of driver corrections and fixesI sort of have a little pressure to buy one soon. The reason is that my brother, who is taking over my old parts, will be wanting his new PC soon. I expect within a month is about it. He wants to upgrade and is not nearly as into it as I am, so he will be very happy with my P5E w/ 4Gb of Mushkin DDR2-800, and a video card. I was going to sell him my 8800GT, which forces me to make a choice soon. So you are thinking (based on what?) that the 4870x2 can hold its own in FSX against my lowly 8800GT? This implies you don't hold Tom's comparison as very valid? Wrong drivers used during the comparison?If you are not personally using or trying the 4870x2 or even 4870, have you read any testimony that these cards are cutting it in FSX? By cutting it, I mean to say can do AT LEAST as well as a 8800GT. As I say, I hate to take a downturn in FSX perf so I can have better IQ, and better perf in other games. I understand there may be AA troubles with the 4800's? What do you know, for sure?

[Guest Nick_N]

Thats just it NoelRight now I have not heard anything about the x2 cards cutting it in FSX like the Ultra or the GTX. I think the card has potential but until a bit more time passes and more driver releases come around, its still a bit of an unknown what it may have the potential to do with FSXOther games, such as Crysis, that card is showing some very impressive AA ability with no loss in frames as compared to the 280 and I really think ATi is focusing on that app due to it being in the spotlight for perf and benchmarking.I know you want that ATi hardware engine though and I do understand where you are coming from about that. If I thought for sure it would kick what you have as least as well as an Ultra (for sure) I would not hesitate to suggest it however I can't so I would rather err on the side of caution. Not to say that may not change next month. Its really early in terms of drivers. I usually wait for at least 90 days from release and then gather data on a card to allow it time to get through its release driver pains... the 280 is just starting to come out of that 90 day phase, the x2 has another 60 days to go... which is why I was hinting to you about waiting a bit longer.I have more confidence in recommending the 280 over the x2 at this point in time as I would not want you to be disappointed