home *** CD-ROM | disk | FTP | other *** search
- Path: news2.cais.com!not-for-mail
- From: swank@cais.com (Robert D. Swank)
- Newsgroups: comp.sys.amiga.applications
- Subject: Re: PGS3.i > BME Virtual Memory ontrol?
- Date: 26 Mar 1996 03:57:51 GMT
- Organization: Capital Area Internet Service info@cais.com 703-448-4470
- Distribution: world
- Message-ID: <4j7pvv$iqu@news2.cais.com>
- References: <4it8sh$bcd@news2.cais.com> <oj.0alj@code1.DIALix.oz.au>
- NNTP-Posting-Host: cais.com
- X-Newsreader: TIN [UNIX 1.3 950515BETA PL0]
-
- Owen James (oj@code1.DIALix.oz.au) wrote:
- : Robert D. Swank (swank@cais.com) wrote:
- : : Wow! Automatic Virtual memory control in BME. Good deal. But, one problem
- : : - I have PageStream and BME located in a partition that has limited
- : : space, so when BME kicks into Virtual memory it fills the partition!
- : :
- : : I can't find anything in the help docs to explain how to direct BME to
- : : use my file area on another partition - anyone know the drill?
- : :
- : : (I surely hope there is one)
- :
- : Me too, because it looks like there's something seriously wrong here. I
- : clipped a small two inch-by-two inch image from a page somebody faxed to me
- : using another utility, then went to BME to try to clean it up a little. The
- : clipped image was only around 30k.
- :
- : When I went to load it, BME ate up the last five megs of RAM I had, then
- : proceded to eat another three megs off my hard drive partition, and even
- : then it only stopped because it ran out of space. Even when it ran out of
- : space it wouldn't let me cancel the operation. I had to reset and manually
- : delete the VM file.
- :
- : Eight megs to load a 30k image? Surely not. Interestingly, I loaded the
- : image directly into Pagestream 3 without a hitch--it only appeared to eat no
- : more than 40k. So what's going on?
- :
- : I haven't tried it with anything else yet, and I doubt I will until we at
- : least get some preferences here to turn the damn thing off, and a Cancel
- : button.
-
- Well, since I posted my original message, I did two things: moved PGS3.0i
- to a LARGE partition, and bought more ram. My file lists as 230k. It is a
- superbitmap 1500x1400x1 two-color image. Large size but 1 bitplane. (I am
- trying to get max resolution for a t-shirt logo)
-
- With 16 megs ram BME took an additional 15megs from the harddrive to load
- the image. My guess is it is working in 24-bit space and needs additional
- space for all its undo levels. I set the undo levels from default 5 to 2,
- which I think helped.
-
- Now, what I was REALLY trying to do is make a vector drawing oout of it.
- BME from 3.0h2 did it with much less overhead, but PGS3.0h2 only
- displayed half the image. The ProVector Demo will load the image in fine,
- at a very large page size. (32x32inch!)
-
- ProVector's tracer module bogged down on the tracing before it go to 90%
- - not good news, but could have been memory limit problem...
-
-
- BME from 3.0i, would start the conversion but since it used up all the
- available CHIP and FAST ram, it appeared to be unable to open requesters
- andcontinue the process. or maybe it appeared to be idle when it was
- actually trying to do something - for half an hour without apparent
- activity and no busy pointer.....
-
- I had the good fortune to get a great deal on a 16 meg simm (DKB board in
- my A3000T) and I tried this again with 30+ megs onboard, and sure enough,
- BME completed the trace in a reasonable time and allowed me to save it.
- However, PGS3.0i still could not display the file properly. It distorts
- the imasge something fierce, to the point it is unrecognisable.
-
- I plan to e-mail my files to Michael with a blow-by-blow, maybe they can
- look into it.
-
- I tried also tracing a smaller more simple image, with better results but
- still not a satisfactory load into PGS3.0i. The documentation appeared a
- little vague as to whether this feature has been fully implemented, but I
- have been waiting for so long.....
-
- Bob
-