The actual thread that explains this is back about message #100 "How fast is a P4 Anyway", the PZ3 file (caled CPU Test) is in Freestuff if you would like to see how your computer stacks up. You load the file and you time the render, that simple. Anyway, here are the results posted by members, sorted. As the question was "How fast is a P4 ..", I would have liked to see more P4 results, we only had two, would have liked to see some more of the faster Celerons, too, I think they make a 1.2 now, don't they? We have plenty of Athlon results, though. Sorted fastest to slowest, faster is better! AMD 1800+ Athlon 512 Mb RAM Win XP Home 037 seconds AMD XP 1800+, 1GB DDR RAM, WinXP 044 seconds AMD 1.4 Ghz Athelon, 768 Mb RAM, Win 2000K - 044 seconds. AMD 1.4 Ghz Athelon 768 Mb RAM, Win 98 048 seconds AMD 1.3 GHz 256meg RAM Win 98 048 seconds Pentium 4 1.7Ghz 384RDRAM WinME 063 seconds Pentium 4 1.9 Ghz 1Gb RAM Win XP 057 seconds AMD XP 1800+ 1 GB DDR Windows 2000 Pro 068 seconds AMD 1700+ (Dual) 1Gb DDR Win XP Pro 073 seconds Pentium 3 850 Ghz 512 Mb RAM Win 98se 081 seconds Pentium 3 1GHz, 512 MB RAM, Windows XP Home 081 seconds. AMD 800 Mhz Athlon 768Mb Ram Win XP Pro 085 seconds Pentium Celeron 600 Mhz 256 Mb RAM Win 2000 Pro 095 seconds AMD 1.0 Ghz Athlon 768Mb RAM Win XP Pro 106 seconds AMD 1..0 Ghz Athelon 512 MB RAM Win98SE 107 seconds AMD 1Gb Athlon, 256Mb Ram, Win 98. 108 seconds AMD ? K6 146 Mb RAM 115 seconds Pentium 3 800MHz, 128MB RAM, Win98 134 seconds Pentium 3- 525Mhz 256 Mb RAM, Win 98 178 seconds Mac G3-500 Mhz 768 Mb RAM ? 215 seconds Mac G3-400 Mhz 256 Mb RAM O.S. 9.0.4, 343 seconds Mac G4 500 Mhz 832 Mb RAM O,S. 9.2.2 476 seconds No, I don't know why there is such a big variance, either.

I did the test on my system, Dual 1.2Ghz Athlon MP, with 1GB of DDR RAM, 64mb Hercules Video Card, Win 2000

Rendered Non-Anti-alias = 17 sec
Rendered w/ Anti-alias = 35 sec

1.0 Ghz Pentium 4 with Radion Video card 90 seconds

Very interesting! P3 1 Ghz 256Mb Ram, Win 98, 70 seconds. Several years ago I tested render time with the little status guy sidestepping across the progress bar. On an older machine (P1, 166 Ghz, 64 MB) he was costing me about 4-5 seconds a minute. I renamed the extension on the little avi file (seems like it may have been named status.avi or something similar) and he disappeared. Poser works fine and my renders are faster, plus I found him annoying. If anyone is interested I can check this out at home tonight and provide more details. Sorry if this old news.

Damn, time to get me a new computer. :D

Bo is correct. The name of the file is "status45.AVI" and can be found in the following folder: C: ... ... Poser4Runtimescripts This is the movie of the little dancing guy that entertains you while your picture is rendering. He also eats up processing power and time so your picture takes longer to render. That is, it renders more slowly. What you can do ... If you change the filename from "status45.AVI" to "status45.AVI" (just "rename" and add an underscore ("") to the beginning of the filename) This makes it easy to find and, also, very easy to change it back if you ever want to do this. Now, when you render, Poser can't find the .AVI file and does not "play" the movie. Your renders will go faster. cheers, dr geep ;=]

Hardware: Dell Inspiron 8100, P3-mobile processor @ 1GHz, 384MB RAM, GeForce2Go 32MB. Software involved: Win2k Pro w/ SP2, Poser4 Pro Pack Active progs running: Poser4 Pro, Netscape 6.2 and the Windows file explorer. TSR's running: Winamp agent, Novell network client, Active Desktop. Time: 132 seconds. /P

Oh, render time w/o the little dancing guy (same conditions) = 123 sec. I shaved about 9 seconds off the render with the .avi disabled. Not bad... /P

I accidentally rendered it both times in a seperate 1024x768 window... was that intentional? /P

Yes, it was, the PZ3 should have automatically called out the render size in a seperate window and turned on a-a and stuff. I had it open in a tiny window to suit all screens. If course it will render faster if you change things, but the real purpose is to see how one CPU compares against another in Poser. I'm sort of suprized at: How good the AMDs 1.3 and over do in Poser (mine is the 1.4 that did 44 and 48 seconds, but many others got similar results). Mine is fast enough that I seldom bother with non-antialiased renders, for example. How poor the Macs do - mine is the one second from last. Especially after all the Mac speed hipe they put out. Don't get me wrong, I like Macs, but they are not a good choice for Poser.

Re. the AMD's - I'm not suprised at all. On most benchmarking and 3rd-party hardware news sites, the benchmarks on the Athlons usually beat the crap out much faster P4 chips, especially in the video-intensive performance departments. Heck, it's been this way since the old AMD K6/2 chips came out. Also, P4's have a horrendous latency problem (though a lot of it should have cleared up after Intel added DDR memory ability to their mobo chipsets... Rambus amplified the latency troubles big-time, and Rambus was all you could get w/ a P4 for the first year or so.) HTH to explain a little as to why the numbers shape up like they do... /P

AMD 900 128 or 256 meg memory, ati raedon 32meg ddr on win 2k time : 62 seconds

165 seconds - PIII 500Mhz, 384mb RAM, Win98SE (no difference with renderman anim disabled). Thanks for hosting the experiment, Mr Burton. It's life, Jim, but not as we know it as a social one. ;-)

Penguinisto wrote: ...Athlons usually beat the crap out much faster P4 chips, especially in the video-intensive performance departments. Video (data handling and processing) is one area where the P4s are consistently particularly good in comparison to the Athlons. The Athlons beat the P4s on raw number crunching (due to their superior floating point performance), and raw number crunching is what Poser makes use of when it is rendering (as does Bryce, Max, etc). The P4s are also good where programs are written to take advantage of their successful features - Lightwave7 is one 3D app that now renders much faster on a P4. Owners of P4s and Athlons (and current Macs) can all be satisfied that they have highly capable computers that make the pro workstations of just two years ago look like steam engines.

Here's another: Athlon 2000XP 1G ram Visiontek GeForce 3 64 Win 2K 24 seonds.

67 Seconds
1GHz Athlon (Old Slot A)
192 MB PC133
Creative GeForce 2MX 32MB
20 GB Maxtor 7200rpm
Windows 2K Pro
No running man
32 Bit color

So my system is as fast as the AMD XP 1800+ 1 GB DDR?

Don't know about these results. I wonder what factors are really at work here.

Glad to see the Athlons stomping everything. Now maybe people will realize I'm not kidding when I say AMD makes a better processor. :) (For sake of interest, I use a P3, the last good Intel cpu. And yes, I'm blunt.)

I dont even use poser but I can't help but chime in here about how I think that AMD is the best. :)

I posted the 37 second score with my Athlon 1800+ and the 81 second score with my P3 850. I just wanted to point out that the Athlon has DDR RAM and both machines have had the AVI side-stepping man turned off for weeks.

BTW - I realized that one thing that affects performance is your monitor specs... For example, I use UXGA instead of VGA (it's a laptop), w/ 119 pixels/inch. I suspect that render times on a standard SVGA monitor would be much faster :) Oh, and PJF: You're right - I had misstated things... I hadn't seen the benchmarks for about 6 months, but a quick check at Tom's Hardware (http://www.tomshardware.com) bears you out. Thank you for the correction. /P

The other strange thing is how many of us seem to be on AMD chips, unless the Pentium 4 guys are holding back ;-) I mention this because all you seem to see advertised is Pentium 4 mahines, or Celerons for the low end. I built my own, I gather most of us did, rather that buy an already set up box. One downside of the faster AMDs though, they are really "hot chips", in both meanings of the word, and they do require a often very loud cooling fan, or several of them. I'd give up a little speed for a little less noise, P4s might be better choices that way. I think the newer AMDs are a little better than the ones like mine, though.

first machine 35 seconds amd 1800 1 gig ram geforce 3 64 megs win xp second machine 1 min 55 seconds celeron 700 128 ram video? unknown but a cheapy win xp

"I built my own, I gather most of us did, rather that buy an already set up box" Well, I don't have the skills to build my own and had to buy what I could get off the shelf. Quite frankly, the relatively small selection of Athlon pcs that were available sucked (Celeron's and Durons were all over the place, but I wasn't interested in those). Small cases, small hard drives, no CDRW (more often than I would have thought), crappy integrated video cards using shared memory, etc. I bought a whole computer, not a processor. The Athlon choices I was given just weren't up to par. It would have cost me a mint to upgrade to match the other features of the P4 I choose (there wasn't a PIII to be found btw...only in laptops). I actually priced it out for the best Athlon I could find and would have had to spend $160 more to make it otherwise comparable...plus all the time and effort to change all the hardware. Then I'd have the joy of having that unused hardware reminding me a big chunk of money was spent for them to sit there and do nothing. Truly not worth it for a few seconds rendering time. I use my PC for lots of things, not just 3D work, and it performs great for everything I do. Anyway, I don't really care if some Athlons "beat" my computer, my computer beat some Athlons too. Big deal--it does what I want it to do and I'm happy so that's all that matters :)

I know this has been mentioned before, but for those who don't know, your video card has absolutely nothing to do with your rendering time as far as Poser is concerned. The cheapest 256kb ISA card (if they ever even made them) should theoretically perform as well as the lastest GeForce 4 in Poser. Everywhere else, a good video card makes a difference but not here. And another thing..... There is a certain satisfaction that comes from building your own computer using components of your own choosing and then seeing it perform at levels well beyond those of the "off the shelf" systems costing twice as much. "Can we build one for you?"....LMAO!

I'm bummed out. My dual 533 G4 Mac was soooooo slow at rendering the file I gave up at 5 minuts!!! Hopefully Poser 5 will support Altivec instructions on the Mac side when it comes out.

And now, for a benchmark at the other end, Whimpypooter (P2/133Mhz/32Mb/W95) spit that puppy out in a swift 1970 seconds! (No kidding - 32 min. & 50 sec.)

Poor, poor whimpypooter..... You must have had the AVI guy on.

Wow, at least I'm not dead last. 615 seconds on an iMac G3/233Mhz/160mb/OS9.2 I'm gonna go turn off my dancin' guy now. I need a beer.

Dell Latitude C600 laptop PIII, 750 MHz 256 Mb RAM Windows 2000 Pro 98 seconds. Not bad considering the fact that I had other applications open as well. I have ordered a 2.2 GHz Pentium 4 machine, guess that will speed things up a bit ;-)

It is actually not that hard to build your own, and yes there is a certain satisfaction in doing it, the only real scary thing is what if it doesn't work the first time you fire it up? If it gets down to you don't know if it is the motherboard or the memory or the graphcs card or the cpu and they all came from different places you have a major problem! I've built all of mine since my 386, though, including three 486s, 3 Pentiums and now an AMD. I only had one that didn't fly the first time, a bad motherboard that I finally convinced the vendor wasn't some kind of a setup problem. Building your own lets you do partial upgrades too, I even upgraded from a AMD 1.1 to a 1.4 before I really got it in service, the chip was only $125 then (more now).

Amazing, removing that silly dancing guy sure made a difference. My rendertime changed from 43 seconds to 37 seconds. (finally the same score as Johnnydnh :-))))

I've built my own computers, for the most part. Yes, you get more control, get the satisfaction etc. But you end up being your own technical support department. If something goes wrong, you're on your own. A year or so ago, I got the bright idea to update the BIOS on my AMD K62-400. Two weeks later I brought it into a local computer shop. To make a long story short, I ended up with an (ugh) Intel Celeron motherboard, with all those nasty builtins (video, network, sound, etc.) I had worked hard to buy the humble parts I had. Those parts were given to me in a box. I could have gotten a brand new, more powerful computer for what I paid to "have this fixed." I haven't had the resources to try rebuilding another computer. Let's not be too quick to recommend that someone go out and build their own computer. Or, at least, let's tell them some of the cold hard realities, and possible pitfalls. Ron

Buying your own computer is great, but it only works when you are able to do your own support. But most people are not. Those people had better buy a Dell or something like that. A compromise is to choose all components of your machine, but to let the local computer store build it for you. Marc

Great idea. If I understand correctly Dell allows you to do just that!

Not quite. I'd like a Athlon XP instead of a P4 :-))) But indeed Dell gives you quite a lot of room to make your own choices.

Pentium 4 1.7 GHz, 256 MB RAM, 67 seconds

Congrats Mjtdevries! What was the fastest Dell computer? Since they're so excellent they should be right at the top of the list right? And if they seem to be a bit slower than our home-built rigs, maybe their fine and indispensable technical support people can enlighten us as to why this is so? Dell has an easy payment plan too...only $90.00 per month for 48 months (with your good credit) and you can get one built for you! Oh, that doesn't include the extended 3 year service plan, but for a few bucks extra, they can roll that into your installment plan. Imagine paying $90.00 a month in that 4th year. It would be like owing over $1,000 on a 4 year old computer! Like P.T. Barnum said "There's a sucker born every minute" None for me thanks.

Dell does not use any AMD processors. So they can't have been at the top of the list because those were Athlon XP 1800+ machines :-))) As I said I before I prefer homebuilt for my own computers too, but if I was going to buy or advice a brand PC, then I'd choose Dell over Compaq, IBM, HP and such. Or the alternative I suggested, if you have a good computer shop nearby.

BTW, to the guy with a 500 Mhz G4 and 800+M RAM: My 300 Mhz G4-enhanced 7500 with 240M RAM rendered the scene in exactly the same time (476 seconds). Just got my 128M stick in the mail this morning, had to test it out.

There is a certain satisfaction that comes from building your own computer using components of your own choosing and then seeing it perform at levels well beyond those of the "off the shelf" systems costing twice as much. "Can we build one for you?"....LMAO

Speaking as someone who built my own systems for years and still do when a client wants something hyper custom I can safely say this...

These days I buy my systems from Dell when I am looking for big iron or eMachines when I am looking for budget. It simply isn't worth the time any more to match the components, track down all the pricing and put it all together and troubleshoot. I like the eMachines BTW - they use off the shelf parts, the price is good and we have 4 in service that have never given us any problem - 24/7/365 for a few years in some cases.

Is there any satisfaction to it? I suppose - but not after a while... it simply gets kind of old :) I think I've built a hundred or more over the years and I have to tell you the thrill is just plain gone.

3 weeks ago we realized here we were going to be WAY short on processor power for our project. I >COULD< have spend 3 days building new boxes and saved some cash up front but I would have lost more money than that in billable time. Instead we hit Dell at 3 a.m. got our credit approval in 10 minutes and had 2 high end boxes pre-configured for DVD writing in 3 days or so... and I could do billable work in the meantime.

When the machines got here they were in service in under an hour - no driver hassles and it all just WORKED. Could I have saved some up front money? Sure. Would it have been useful? Not by a long shot.

The credit? Yeah, the rate is steep... but we will pay them off in 15 days or so when we bill this client for the time and the extra work we can get done. A short term instant business loan in my books :)

Most of my clients run Dell's as well if they are in production environments. When and IF something dies they replacement part ships in 12 hours usually if they need it fast. They save money on downtime as well.

So, if you have the time to spend on the homebuilding it's a fun hobby in my mind... but when you're doing the dance of billable hours I put it away with most of my hobbies.

For similar reasons we run Intel chips here. Yeah, the AMD is a good chip and has some speed in some cases, but on optimized software we are VERY happy with our P4's and P3's. When we need to encode video (and we do a lot of that) then it is a godsend to have the speed. Oh, and the Intel chips don't catch fire if the fan falls off :)

My results:

1.8 ghz P4 with 512mg RDRAM.

My poser times sucked. Of course I didn't clean boot the box... I have too much stuff happening. Task Manager shows 53 processes running including Outlook and Photoshop. It took Poser 1:53 seconds or 113 seconds total. Beats the heck out of me why. I suspect some oddities in Poser  and this task load.

As a side note: 3D Studio Max ran the test (I matched the lights as well as I could and turned shadows on) in 11 seconds flat on the same machine with the same task load (right after I closed Poser).

So I'll be rendering in Max from now on :)

Soulhuntre, I had my rig put together and running within 4 hours so its doubtfull you're saving all that much time with Dell or e-machines setup. I would be interested in seeing a screen shot of your task manager displaying the 53 items that are running in the background. Based on your render time, it seems that you would have been perfectly happy with the performance of a properly configured PIII 450. But then again, you would lose the certain satisfaction that comes from spending $2,500.00 on a new computer. You also mentioned e-machines. Aren't those throw-away boxes in that they can't be upgraded? You recommend them? I noticed that Dell and e-machines are underscored and highlighted in your post. Are those direct links to their advertising pages? To be fair, you should post a link there to take them here to see the results of our tests. I just clicked on them....they ARE links! How much are they paying you?

Girls, take your pillow fight somewhere else.

Hmmm... a little hostility there? I am not sure what you're so upset about...

Anyway, taking your points in order...

"I had my rig put together and running within 4 hours so its doubtfull you're saving all that much time with Dell or e-machines setup."

Did that include selecting the pieces, picking or ordering the parts, the assembly and load out itself and some sort of burn in testing? If so your faster than I am and more power to you :)

"I would be interested in seeing a screen shot of your task manager displaying the 53 items that are running in the background."

I said processes. If you look at yours you will find a bunch there as well... there is a distinct difference between user programs and processes. But you knew that :) Usually I have 8-10 IE windows, Outlook, Photoshop and Max or Poser.

"Based on your render time, it seems that you would have been perfectly happy with the performance of a properly configured PIII 450."

I have a few of those, and I sure wasn't happy I promise. The render times on them aren't what I need for the work we do... our current project needs more than a  hundred thousand frames of animation. The most important thing this becnhmark pointed out was how invalid Poser seems to be as a benchmark - look at the variations in the results and you will see that raw CPU horsepower seems to be far from the only factor. I am sure if I shut everything else down it would be better - but I don't really have the time (in outlook alone I have 15 pending email replies in progress... if I close Outlook I have to save them all then remember to open 'em back up etc.).

"But then again, you would lose the certain satisfaction that comes from spending $2,500.00 on a new computer."

I don't have any. My personal satisfaction doesn't come from the box or how much I paid for it. My satisfaction comes from what the box does for me, and whether I get it all done on time and at a profit. The boxes I buy and recommend do that... that is satisfying I suppose. If I wanted to draw personal satisfaction from the things I own there are better options than my PC, I change it every 6 months or so anyway.

"You also mentioned e-machines. Aren't those throw-away boxes in that they can't be upgraded?"

Well that depends on how you mean. We sure don't throw them away - they see service as web servers, file servers and render farm boxes. Some run broadcasters for streaming video and audio and so on. One is a dedicated SQL server box.

The thing I like about eMachines is that they use standard parts in fairly standard boxes. I haven't needed an upgrade I couldn't; put in. Video card swaps, adding HD, adding CD-ROM, upgrading sound and so on. If I wanted to replace the MB I could probably do it, worst case is I have to buy a new case and transfer the parts.

Then again, we don't upgrade like that. When a machine is too slow for front line production we put it in service on the second line as a server and so on. That's one of the reasons we have 5 systems in service now without having a huge budget. Not everything needs a front line box.

"You recommend them?"

Yup, when they are appropriate. I have clients, friends and relatives with them. I own them myself. it's a good inexpensive box from a good company. Why wouldn't I recommend them?

I would not recommend a brand I have had trouble with. Compaq comes to mind. Proprietary components, no OS upgrade path and missing drivers for newer OS's... bad tech support and odd configurations. Not me buddy.

"I noticed that Dell and e-machines are underscored and highlighted in your post"

The term is "link".

"Are those direct links to their advertising pages?"

No, those are direct links to the home pages of the companies I mentioned. Yes, those pages have advertising... they also have technical information and forums where you can see what other people say about them. The beauty of the web is that when you mention something you can link in other sources of information so the people reading your text can do some checking on their own :)

"To be fair, you should post a link there to take them here to see the results of our tests."

I suppose I could, except I am not sure what the point would be and I am not active in those forums. Why don't you go do it if it upsets you so much?

"I just clicked on them....they ARE links!"

Glad to hear it! I went to all the trouble to make them links so it's good to see that it all works the way it is supposed to. I hope you weren't too surprised.

"How much are they paying you?"

    Millions. in fact I don't have a regular job... Dell pays me millions of dollars so that I can sell a few more machines for them by posting good things bout them. It's all a plot to trap you. Curses, you foiled the plan!

Seriously, I am not sure what you seem so torqued about. I just posted my experiences with the tools I use. Yeah, I said good things about a vendor I like - I do that with anything I think deserves my support. If it is good I let folks know when it is relevant. Ah well, not worth fighting about.

"The most important thing this becnhmark pointed out was how invalid Poser seems to be as a benchmark - look at the variations in the results and you will see that raw CPU horsepower seems to be far from the only factor" Ha! That is one way to look at it, another is sometimes real world results don't always meet are expectations! I wish we had some results for Celeron 1200s or 1300s though, I bet they would stack up pretty well against the more expensive Pentium 4s. I wonder how many of the store bought machines use DMA 99 hard drives that are actually running as such, for example. Anyway, here is most of the above added to the list. ******************************************* CPU, Speed, Amount of RAM, Operating System, time AMD 2000XP 1G RAM Win 2K 024 seconds AMD 1800 Mhz 1GB Ram Win XP 035 seconds AMD 1800+ Athlon 512 Mb RAM Win XP Home 037 seconds AMD XP 1800+, 1GB DDR RAM, WinXP 044 seconds AMD 1.4 Ghz Athlon, 768 Mb RAM, Win 2000K - 044 seconds. AMD 1.4 Ghz Athlon 768 Mb RAM, Win 98 048 seconds AMD 1.3 GHz 256meg RAM Win 98 048 seconds Pentium 4 1.7Ghz 384RDRAM WinME 063 seconds Pentium 4 1.9 Ghz 1Gb RAM Win XP 057 seconds Pentium 4 1.7 GHz, 256 MB RAM, 067 seconds AMD XP 1800+ 1 GB DDR Windows 2000 Pro 068 seconds Pentium 3 1 Ghz 256Mb Ram, Win 98, 070 seconds. AMD 1700+ (Dual) 1Gb DDR Win XP Pro 073 seconds Pentium 3 850 Ghz 512 Mb RAM Win 98se 081 seconds Pentium 3 1GHz, 512 MB RAM, Windows XP Home 081 seconds. AMD 800 Mhz Athlon 768Mb Ram Win XP Pro 085 seconds Pentium 4 1.0 Ghz ? ? 090 seconds Pentium Celeron 600 Mhz 256 Mb RAM Win 2000 Pro 095 seconds Celeron 700mhz 128 MB RAM Win XP 105 seconds AMD 1.0 Ghz Athlon 768Mb RAM Win XP Pro 106 seconds AMD 1..0 Ghz Athelon 512 MB RAM Win98SE 107 seconds AMD 1Gb Athlon, 256Mb Ram, Win 98. 108 seconds AMD ? K6 146 Mb RAM 115 seconds Pentium 3 800MHz, 128MB RAM, Win98 134 seconds Pentium III 500Mhz, 384mb RAM, Win98SE 165 seconds Pentium 3- 525Mhz 256 Mb RAM, Win 98 178 seconds Mac G3-500 Mhz 768 Mb RAM ? 215 seconds Mac G3-400 Mhz 256 Mb RAM O.S. 9.0.4, 343 seconds Mac G4 500 Mhz 832 Mb RAM O,S. 9.2.2 476 seconds iMac G3 233Mhz/ 160mb RAM OS9.2 615 seconds

"Ha! That is one way to look at it, another is sometimes real world results don't always meet are expectations!" A possability :) The thing is, we knwo that a single test case (poser) only tells us one thing - how fast poser runs on any system. In and of itself that is useful, but it is clearly tied to much more than simply CPU speed. This machine benchmarks right where it is supposed to for 3DS Max, photoshop and Premier. So P4 aware software seems to be doing exactly what it is suppose to. ::shrugs:: I'll be damned if I understand it :) Liek I said, seeing that Max can runt he same scene in < 12 seconds tells me the problem is poser... not the box itself :)

So the answer to the original post is: If you want the fastest render times in Poser, get an AMD. If you want to have 20 or 50 or 75 programs open at once and render more slowly in Poser then by all means by a P4. Don't buy just any P4 however, make sure that it is a Dell P4 that has the sticker "Intel Inside" Another thing to think about is that I also posted the P3 850 score of 81 seconds and that machine is not a Dell either. The problem may not be Poser....it may be your Dell.

I don't want to turn this into a bash anything thread, but Tom's Hardware has always said the P4s were turkeys, ;-) Intels pushing them has a lot more to do with hidden agendas than any other CPU ever sold, I think. If I was going to by a Intel system I'd go with one of the faster Celerons, AMDs are sort of a pain, too, true, for the noisy cooling fans! As to why the Macs are so slow is another story, I really think there is a problem with Posers code. I'd guess earlier versions of Poser were actually done for the Mac, and ported to the PC (some of the files have .TIFF as an extension, which is a clue!), perhaps it is done the other way now, and not very well.

Jim Burton, Glad to see you mention Tom's Hardware Guide. I did most of my research on building my computer there. As far as the noisy cooling fans go, your are absolutely right. I have 4 of them in my rig and can hear them from 3 rooms away. If I was going to recommend an Intel CPU, I would have to go with a PIII 1.0 or 1.2 ghz model. The PIII appears to have a significant performance advantage over the Celerons unless things have changed very recently. I would stay away from the P4 due to their high cost which is not justified by their performance. Tom did a nice article on his site about "Best Bang For The Buck CPUs" which I thought was very enlightening.

A developer who will go nameless here has said that Poser is traditionally coded in the CodeWarrior IDE for Mac then ported to PC. In the bid for compatibility and identical results, the code in all likelihood does not have any optimization for any hardware or OS hooks, leaving the efficiency up to the strength of your system. Judging from the hints kupa's been dropping about conforming clothes, gravity and strand-based hair, I'm guessing that Poser 5 is being optimized for both MacOS and Windows to make these features even plausible. I'm not an expert in these things, but I recall Blender's authors being very reliant on open graphics standards like OpenGL to allow them to port from Windows to PocketPC to Linux to OSX. Apparently MacOS 9.x wasn't as OpenGL-friendly. When Curious stops supporting MacOS 9 and its legacy issues and starts reaping the benefits of OSX's graphics engines, I think Mac rendering times will substantially improve. Curious' response along these lines is essentially "we'll wait and see."

"I would stay away from the P4 due to their high cost which is not justified by their performance. Tom did a nice article on his site about "Best Bang For The Buck CPUs" which I thought was very enlightening" And for most general use this is currently true, no disagreement at all :) However for code that DOES understand the P4 the upgreade is dramatic and useful for us. 3DS Max and our Mpeg encoder's are an example of this. Photoshop is starting to incorporate them as well. For what I use it for the P4 is dramatically faster than the P3... right now. I can wait the time needed for other applications to catch up and optimize for the P4... Word is plenty fast anyway. Intel is "pushing" the P4, no doubt ... but Intel is also full of really intelligent folks and while it will take time for the optimizations to become the norm they will come :) My interest in the Mac results was more that their numbers varied by much more than the processor speed. That is interesting :) Ah well, I do want to say thanks to Jim for the test, it showed me how much more profitable my time is using the Max system as my primary render tool... I had been bringing my poser scenes in only at the end of their life - now I will pull them in much earlier :)

I use Max a zillion hours a week, but it is all for modeling, I've never done a single real render in it. The only time I tried I ran into problems using my hair, I think you need to turn on double sided or something, otherwise the backside polys don't get rendered. I aught to get up to speed in it, another thing for the to-do list. However, my system is fast enough in Poser it wouldn't be primarily for saving time, but it would be nice to not have to fake the reflections like I do now. ;-) Incidently, the file is still in Freestuff, search on my name. I'd still like to get a couple more resulys, especially 1200+ Celerons or some 800 mhz+ Mac G4s. My school is updating my Mac G3 to a G4 800, but It will probably not be here to mid summer. I'm (eventually) going to add a text file with the results to the zip, in a couple years we can rerun the test and say "What did we think was so fast about these computers..." !