May 2015 - Fractint - mailman.xmission.com

Re: [Fractint] Fractint Digest, Vol 147, Issue 9
by JackOfTradeZ＠comcast.net 16 May '15

16 May '15

. ""What about the issue of the graphics card (or onboard chip) needing to support the DOS VESA video modes used by Fractint?"" Any P2 or P3 box as is (as was) will have a graphics card from that era, they all work(ed) fine for me - never had a problem. It seems that the older cards actually give better video options with FractInt than the newer ones - I have noticed many times. Obviously new cards are better in all respects, I think it is a DOS/compatibility issue. I just bought a state-of-the-art video card (circa 2003) for $10 and it gives me 1600 res option. As for chips - I have determined that Athlon64 works better than Intel P4. P2 and P3 work the best of all. .

2 1

Re: [Fractint] Mandel Machine (Roger Kaufman)
by Roger Kaufman 14 May '15

14 May '15

Hi Richard, > ./xfractint @test1.par/VsionOfaDistantSea This par, which contains the formula core dumps. When I used "Poachers_Afoot" which uses a system formula it works. ./xfractint "@test2.par/Poachers_Afoot" > I can send you my scripts that run from a cron job if you want. The > relevant part is this: > > bin/xfractint -disk -geometry $2x$3 \ > batch=yes savename=$stream.gif "@$stream.par/tmp" > > The batch=yes makes it render the image and then exit immediately > after it is done. I am having trouble when the geometry gets top big. It immediately writes a blank gif and exits. I don't have time to debug now but I could put in debug statements to see where it fails. It may be a problem due to the fact I am running in a virtual machine instead of natively. Interesting, once (but only once) this command worked by generating the image of 1600x1200. ./xfractint -disk -geometry 1600x1200 batch=yes savename=test.gif "@test2.par/Poachers_Afoot" How is geometry supposed to work. If I tried 800x800 it instead generated an image which was 1068x800. Maybe my geometry isn't legal? Roger

2 1

Re: [Fractint] 91 year old man wants fractint help!
by Timothy Wegner 14 May '15

14 May '15

For everyone's information , Rivka (daughter of the Fractint fan mentioned in the quoted message below) is now on this list. So between the list and one of the list members who is helping off-list, maybe we can get Rivka's father set up. Rivka, don't hesitate to ask questions here or report progress! Tim On Tue, May 12, 2015 at 7:25 PM, Timothy Wegner <tim(a)tswegner.net> wrote: > I just got the following email, which I found very touching: > > ================= > Dear Mr. Wegner, > > I am writing on behalf of of my 91-year-old father. He has been using > Fractint for decades to create art like the attached. > > Unfortunately, computers breakdown and we have been trying, > unsuccessfully, to create a vintage/legacy system that will run the DOS > version of Fractint. > > I recently purchased a vintage Dell laptop running windows 95 from eBay. > It is not working for him. My father seems to remember a command using the > keystroke B which removes the color from the fractal and only leaves > an outline. He then imports this image into Windows Paint in order to > resize it for the canvas. We cannot find that functionality now. > > So, my questions to you are: can you recommend the components of a system > that we could put together to get my father back in business using the DOS > version of Fractint? Do you know of a command in the program that removes > the color? And finally, are you the best person for us to consult with on > this, or can you recommend someone? > > On behalf of my father and my four siblings who he has trying to tackle > this problem, I thank you for your time. > > Sincerely, > Rivka Morgan-Sherman > ======================== > > I am telling him that a Win95 computer should work fine. I am not clear > from his email why Win95 is not working for him. However, I am not sure > what the "B" command is he is referring to. Any ideas? > > I am inviting him to join the list. I am willing to help but I suspect > some of you who are currently using fractint might be better. Also, if any > of you would like to correspond with this gentleman, just let me know > off-list. > > Tim >

1 0

91 YO FractalNaut
by JackOfTradeZ＠comcast.net 13 May '15

13 May '15

. Do what I do, what I suggest many times to people on this list with the same problem, but no one ever does it. Purchase a legacy system, either locally at one of the (many) brick and mortar PC repair outfits, or online (search - there are so many). Get a Pentium-3 at ~700-900 MHZ - they work the best. You shouldn't shell out more than 100-200 $ if that; sometimes much less even. Install DOS, or Win9X ; use a fresh HD or simply format the one that came with it. Run FractInt and have fun - just like the good old days. If you get to Win9X then I can tell you how to configure it so it boots to DOS and STOPS. If using just DOS you can transfer files with a USB Mem stick. In BIOS, set the flag for "legacy USB support" and it will work! Since DOS is no longer supported etc., I think it is legal to share the installation files. I can post the original DOS 6.22 files [3 floppies] on my server; you can download and use to install. Can't do with Win9X. Although I have the Disks for 95 and 98. .

2 1

91 year old man wants fractint help!
by Timothy Wegner 13 May '15

13 May '15

I just got the following email, which I found very touching: ================= Dear Mr. Wegner, I am writing on behalf of of my 91-year-old father. He has been using Fractint for decades to create art like the attached. Unfortunately, computers breakdown and we have been trying, unsuccessfully, to create a vintage/legacy system that will run the DOS version of Fractint. I recently purchased a vintage Dell laptop running windows 95 from eBay. It is not working for him. My father seems to remember a command using the keystroke B which removes the color from the fractal and only leaves an outline. He then imports this image into Windows Paint in order to resize it for the canvas. We cannot find that functionality now. So, my questions to you are: can you recommend the components of a system that we could put together to get my father back in business using the DOS version of Fractint? Do you know of a command in the program that removes the color? And finally, are you the best person for us to consult with on this, or can you recommend someone? On behalf of my father and my four siblings who he has trying to tackle this problem, I thank you for your time. Sincerely, Rivka Morgan-Sherman ======================== I am telling him that a Win95 computer should work fine. I am not clear from his email why Win95 is not working for him. However, I am not sure what the "B" command is he is referring to. Any ideas? I am inviting him to join the list. I am willing to help but I suspect some of you who are currently using fractint might be better. Also, if any of you would like to correspond with this gentleman, just let me know off-list. Tim

2 1

Re: [Fractint] Mandel Machine (Roger Kaufman)
by Roger Kaufman 12 May '15

12 May '15

Hi Richard, > My pages render the images at 1600x1200 with 3x oversampling, so the > actual rendered dimensions are 4800x3600. My system has this processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 60 model name : Intel(R) Core(TM) i7-4770K CPU @ 3.50GHz stepping : 3 microcode : 0x19 cpu MHz : 3506.459 cache size : 6144 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 2 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 5 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl pni ssse3 lahf_lm bogomips : 7012.91 clflush size : 64 cache_alignment : 64 address sizes : 39 bits physical, 48 bits virtual power management: I have been trying to load the par in disk video mode and it core dumps right away. The planned command is ./xfractint -disk @test.par/VsionOfaDistantSea -geometry 4800x3600 Trying this with the default fractal does not core dump, but at such a large window, how do you get the dialog that it is done? I just guessed at it as I assumed it was done right away. I pressed 's' to save it and then escape and 'y' to exit. Might there be a way of specifying the output file on the command line? Do you simply resize the image with an image editor afterwards?

2 1

Re: [Fractint] Mandel Machine (Roger Kaufman)
by Roger Kaufman 11 May '15

11 May '15

> I will try to find one of Jim Muths fractals where he documents the generation time. I will try it under Dosbox then under the Linux version after understanding the local environment. I have some learning to do so this may be a while. I tried the first fractal on this page... View of a Distant Sea. The time given on this image was 4+ hours although they don't give what environment this is. That kind of time is reliably assumed to be on a Dos system in some way. http://user.xmission.com/~legalize/fractals/fotd/2014/11/index.html I started it in DosBox configuration quite some time ago and it is almost done. Estimated time will be about 3 hours. Meanwhile over in Ubuntu Linux in the Oracle virtual machine (Windows 8 host, Ubuntu 64 guest), it took 2 minutes and 11 seconds. I ran it twice and it was within milliseconds of 2:11 both times. I compiled the latest xfractint and was running it in a local directory. Resolution was 800x600x256 in both tests. Just for the record my relevant hardware. This machine was built about 1 year ago. Asus Z87-A DDR3 1600 LGA 1150 Motherboard Intel Core i7-4770K Quad-Core Desktop Processor Corsair Vengeance Pro 32GB 4 x 8GB DDR3 2400MHz C11 Memory Kit Cooler Master Hyper 212 EVO XFX CORE Edition HD7850 860MHz 2GB DDR5 2x Note that at no time did fans spin up. I did some hunting to try to find other ways to generate the image. I tried ChaosPro which wasn't very compatible. Once I got the par and formula in (I had to edit) I got a blank image. The program hasn't been updated since 2010 so it seems abandoned. Uninstalled. I don't want to buy or try Ultrafractal. It also hasn't been updated in years. That being said, there are no freeware fractint compatible fractal programs. I had said that Fractint was legacy, but now that I see the work having been done on xfractint I take that back. Its the only way to go.

2 1

Re: [Fractint] Mandel Machine (Roger Kaufman)
by JackOfTradeZ＠comcast.net 10 May '15

10 May '15

. You got it! I discovered long ago that WHERE you zoom in makes a BIG difference in speed; the "chaotic" areas clearly take more time (iterations) since that is needed to distinguish the points and pixels. Obviously, the closer you are to a "lake" area the greater the chaos. Note that many deep zooms on the web are zooming into the west needle ..... [I am guilty of that also <G>]. Now waiting for a truly refined and debugged XFractInt to give me the final reason I need to get into Linux ... .

1 0

Re: [Fractint] Mandel Machine
by Roger Kaufman 09 May '15

09 May '15

On 5/9/2015 2:00 PM, fractint-request(a)mailman.xmission.com wrote: > I would like to give Mandel Machine a try, but I don't have 64bit WinDoZe, and it looks like MM doesn't have a 32bit version. > > I would try it near the point where the 2 main bulbs intersect. It is very difficult to access with FractInt or any other fractal program. I did one zoom-in to that point and it took forever to get even to E+12. > > Also, MM says it has an iteration limit of 400 million - it needs to do better than that for infinite zooms. FractInt still wins with it's 2.1 BILLion max iter! MM seems to be nice and speedy but maps are a nuisance. You can import most images (literately pictures) and form a map by sampling the colors from it. The closest I could get to Fractint's default map was to capture the image of the default scene and sample that and it wasn't very good. I wonder if the MM developer could include the default Fractint map in the list? He has Ultrafractal's default map so it seems only fair. MM breaks down at being speedy in areas around the bulbs with the background detritus (I think Jim Muth had a better description here... high iteration background stuff?). I believe for the MM deep zoom video they hung around Scepter Valley were there wasn't so much of this. It seems even with the new math, if all the pixels in a scene are chaotic, it takes similarly as long to calculate the image as just plain iterating. Fractint is still king imho. I have been a long time user of Ubuntu Linux run under the freeware Oracle VM Virtualbox. I was just looking at xfractint package. While it needs some work, I am truly impressed. The performance is simply super. The display had a refresh problem but moving the window a tad refreshed it (this might be a problem caused by running under the VM). Many fractals were already done in just a few seconds. The keyboard menu was nearly the same. If I new the directory structure for the local files I would relocate what I have from dosbox over to Linux. There are differences in the Linux world that don't make sense from the Dos world. e.g. to change resolution, under the delete key it would (ideally) just change the windows size nicely for you. I believe that it could even have the color rotation by refreshing the image after each color change. (I've seen this work under GLUT programming but I am getting into too much tech). If a stone soup approach could be made to develop Fractint on Linux it would be one super program. It already is in many ways, it just needs a little honing. Seeing what I have seen it should the primary platform for development. I will try to find one of Jim Muths fractals where he documents the generation time. I will try it under Dosbox then under the Linux version after understanding the local environment. I have some learning to do so this may be a while.

1 0

Mandel Machine
by JackOfTradeZ＠comcast.net 09 May '15

09 May '15

. I would like to give Mandel Machine a try, but I don't have 64bit WinDoZe, and it looks like MM doesn't have a 32bit version. I would try it near the point where the 2 main bulbs intersect. It is very difficult to access with FractInt or any other fractal program. I did one zoom-in to that point and it took forever to get even to E+12. Also, MM says it has an iteration limit of 400 million - it needs to do better than that for infinite zooms. FractInt still wins with it's 2.1 BILLion max iter! .

1 0