You need to tell the makers of Focus Magic about this, it’s almost certainly a bug in their plugin.
Dear John,
I am having similar problems with Focus Magic v3.0 in PS CS w/ 16bit images. The only difference is that my error is the following:
===============
Runtime Error!
Program: C:\Program Files\Adobe\Photoshop CS\
R6025
– pure virtual function call
===============
I have tried reinistalling both PS-CS and Focus Magic, but I still get this error EVERY time upon completion of applyng focus magic. It’s weird how it happens just after the plugin is applied (I can see the changes appear, then the error message pops up and I lose all my work and Photoshop shuts down).
Any help in this matter would be greatly appreciated. I have contacted Focus Magic about this and they say that no other users have had problems with PS-CS except me, although from this thread I can see that this is clearly not the case.
I am also getting the R6025 -pure virtual function call error message from time to time. Are you only getting the error messages with 16 bit images? My images are also 16 bit. I am going to send an email referencing this thread. Fortunately, I just purchased Focus Magic, so if the problem isn’t acknowledged and fixed I can ask for a refund (or refuse to pay the charge on my credit card if it isn’t granted).
–John
John,
While I’m not on a PC all that suitable resource-wise for PSCS use, I followed your example and also obtained an error, although in my case it was a "could not complete your request due to a program error" message. I used the FocusMagic 3.0 demo. The PC is a Windows 2000 system.
Regards,
Daryl
I get one of three errors:
1) "could not complete your request due to a program error" (as you received)
2) "runtime error"
3) "abnormal program termination"
The later two cases cause photoshop to shut down and I lose all of my work. I have not yet studied this bug enough to figure out when each of these occurs, as it appears that any of them can randomly result from using focus magic and PSCS.
I am also using Windows 2000. My system has copious amounts of RAM and scratch disk space available. Curiously, one thing that I’ve found lessens the tendency for these errors is to delete your preferences file, as described in the photoshop FAQ in this forum. This is *definitely* not a solution though, just a temporary workaround if I save all the time– the errors still occur, just less frequently.
I just saw the "Pure Virtual Function Call" Photoshop crash for the first time after using FocusMagic. I have not yet been able to reliably reproduce it, but I will also report it to the FocusMagic folks. Perhaps once they see multiple reports they will investigate.
Mathew
I get a "Photoshop made an error and needs to shut down"-error when using Focus Magic. (FM is the cause, not PS)
The bug (in FM) is connected with 16 bit/channel use.
Rob
Thanks to all who replied. The more people who complain/report the bugs to Acclaim Software, the more likely it is that the bugs will be found and fixed. Makes one wonder about whether any beta testing was done, though. These problems are pretty obvious when working with 16 bit images.
–John
I just got a reply from Acclaim support saying that there seems to be a problem with large files in FocusMagic, and that they are looking into it.
Mathew
OK , this tells me that Focus Magic is probably corrupting memory — which causes it or Photoshop to give errors shortly after using the plugin.
I just got a reply from Acclaim support saying that there seems to be a problem with large files in FocusMagic, and that they are looking into it.
I wish they would be as communicative with me. I don’t mind bugs in software, but the manner in which they are acknowledged and fixed is crucial. All I got was a terse reply that they would look into it. I don’t consider a 36 MB file to be "large" (perhaps a 300-500 MB file would qualify for that designation).
I think that the Plug-in is definitely corrupting memory, because the type of error seems to vary a lot. Photoshop can’t save preferences, or is out of RAM, or some other nonsense.
–John
See Ervin Little "Genuine Fractals 3.5. 16bit." 5/11/04 11:59pm </cgi-bin/webx?50>
Pierre,
Maybe I am being dense, but I don’t see the connection between the current Focus Magic bug and the new upgrade to Genuine Fractals 3.5 for 16-bit graphics. Are you suggesting the new Genuine Fractals has a bug?
— Burton —
ditto 🙂 please elaborate… (unless both of us are being dense)
huh, sorry, I misread the titles… the close versions mislead me, I think…
I did the same thing pierre! 🙂
Acclaim Software has told me that they are unable to duplicate the problem on their end, and to please zip up the image and send the list of steps to them. For anyone else who is running into this memory corruption, please report it to them, send images, etc.
Thanks.
–John
Focus magic has told me they found the error and that it is Photoshop’s fault, not theirs. Here is their response:
=========
Hi Sean
After Focus Magic has sharpened the image and passed it back to Photoshop it is up to Photoshop to save the TIFF file correctly. I think you need to contact Photoshop
All the best
Eric
===========
Regardless of the source of fault, I just hope that someone is working on this and that it does not turn into a game of pointing fingers. Hopefully this will get resolved.
Gotta love finger pointing. I can make it crash without saving to TIFF (I don’t use TIFF for my photographs) — just about any image operation after using Focus Magic in 16-bit mode can cause a crash.
Mathew
Two hypothetical scenarios (YOU choose the most likely):
1) Although other plug-ins work just fine with Photoshop CS for 16 bit images, and Adobe has extensively beta tested Photoshop CS before releasing it, Acclaim Software has managed to uncover an obscure bug in the Photoshop plug-in interface which by chance only affects images processed with their plug-in. They report the bug to Adobe Technical Support, who confirm the bug and put it on the list of bugs to be fixed in version 8.01.
2) Photoshop CS is stable, but can have memory corrupted by unruly plug-ins. The Focus Magic plug-in has a bug which causes memory corruption, and the specific error and when it occurs depends on what part of Photoshop memory is corrupted. There is no established beta program to uncover these errors, and technical support gives different answers to different people reporting the same basic problem. The technical support person is not the programmer, who remains unnamed and does not communicate directly with those reporting errors.
Which version is most plausible?
–John
Here is a revised method of crashing Photoshop using the Focus Magic Plug-in which does not require a specific file:
1) Start Photoshop
2) Create a new file 2052 X 2838 pixels, Resolution 300, 16 bit, tagged as Adobe 1998
3) Select the gradient tool and pick a linear, white-to-black gradient
4) Create a vertical black-to-white gradient, holding the shift key down while making the gradient so that the gradient is completely vertical
5) Drag the layer to the copy symbol on the layers palette to create a copy
6) Run Focus Magic Plug-in on this newly created copy of the background layer using blur width 1, amount 100%
7) Save the file as a TIFF with the name temp3.tif
8) Delete the extra layer you just created and "focused"
9) Save the file again
10) If you haven’t gotten any errors yet, repeat steps 5-9 until you get an error
Please let me know if this method will crash Photoshop on your machine–it certainly does on mine.
–John
How is Focus Magic corrupting memory a fault of Photoshop?
2) is far more plausable.
How is Focus Magic corrupting memory a fault of Photoshop?
It isn’t, of course. The "scenarios" were meant to be a tongue-in-cheek 🙂 attempt to make the point that a bug in Photoshop is not very likely given the symptoms, the fact that the problem only occurs with this particular plug-in, and the extensive testing Photoshop has had.
By the way, the "technical support" and "programmer" referred to in scenario #2 were from Acclaim Software, not Adobe.
–John
And that question was rhetorical.
Here is the latest answer from Acclaim Software:
"Thanks for your email. This is almost definitely a Photoshop problem because Focus Magic is no longer active when the problem occurs. The only possible way Focus Magic can cause this problem is if it passes garbage onto Photoshop, but if this was the case you would see it in the image. I think you need to contact Photoshop with the steps you have shown to duplicate the problem."
They conveniently neglect to mention whether they have been able to reproduce the problem on their computer with my list of steps. Don’t you think they would want to report this "bug" in Photoshop to Adobe so that Photoshop CS can be "fixed" to work with their plug-in?
Any comments? Chris Cox, can you suggest how I should go about getting through to them? Almost ready to insist on that refund (or refuse to pay credit card charge when it comes if refund is refused).
–John
Their plugin corrupts memory while they’re running, then that corrupted memory causes random problems later. It is still their bug.
It sounds like they could use some lessons in basic logic and debugging.
Yes, it would also help if they contacted Adobe about this – then I could explain their bug to them in detail. 😉
After a bit more experimentation, it appears that three ingredients are necessary to reliably cause an access violation or other error (on my system):
1) 16 bit image
2) Layered file, with Focus Magic operating on a layer other than the background
3) File dimensions are important, e.g., don’t get the problem with a 2000 X 3000 pixel file, but do with a 2052 X 2838 pixel file
–John
Because this is a memory corruption issue, I suspect that the conditions vary from system to system. This is why it is so hard to reproduce.
I just sent Acclaim a 16-bit .psd file with which I can reproduce the problem on my system — it’s a 2240×1680 image, but it has no layers.
Regards,
Mathew
I am sure you are right about the conditions varying from system to system, but there are a few important clues I ran across:
1) The contents of the file don’t seem to matter from what I can tell. It seems to be the SIZE of the file that makes a difference. For example, I had a file which produced the problem reliably, size 2052 X 2838. I created a black-to-white gradient in a new file of the same size, applied the same blur width and amount, and also reproduced the error.
2) While experimenting with the file size, I discovered that on my system, reducing the width from 2052 to 2047 eliminated the problem, but when the width of the file was 2048, the problem returned. Hmmmmmm… where have I seen that number (2048) before???? Do you think it was just a coincidence that a file 2^11 pixels wide caused the problem, while one less than that didn’t?
3) I also discovered, that on my system a file that was 2047 X 2631 worked fine, but increasing the width to 2048 or the height to 2632
( = 2048 + 512 + 64 + 8 — OK, I know that may be stretching things a bit) caused the bug to return.
Just some food for thought.
–John
Nothing really useful to add but I have NOT had any problems with it on files up to 80 meg (16 bit) on Win-XP-Home 3/4 gig chip memory.
The first post mentioned Win-2K. I also saw that flavor Windows being mentioned on down in the thread. Could it be a Windows-2000 incompatability and not a CS + Focus Magic problem??
Just for interest, are any of you following this and having problems running XP-Home?
Bill
Bill,
I have Windows XP (Home Edition) and have the problem.
Mathew
Well, that seems to rule that out.
Bill
Could you give them Chris’ email address?
Could you give them Chris’ email address?
They finally asked for it, and I gave it to them.
–John
I don’t know about that plugin, you can do it within Photoshop with already standard filters.
Bart,
I don’t know about that plugin, you can do it within Photoshop with already standard filters.
Well, you are right about not knowing about that plugin, or you wouldn’t make the claim that you can "do it" within Photoshop with already standard filters.
It is commonly believed that if an image is taken out-of-focus by the camera that you can’t re-focus it in software, but Focus Magic uses a computationally intensive deconvolution algorithm to, within limits, do that very thing and reverse the process (each point of light becomes a halo) by which an image is defocused.
Because of camera and subject motion and imperfect function of my camera’s and camcorder’s auto focus features, almost all of my images are at least slightly out of focus, so I routinely sharpen their focus in Focus Magic.
Focus Magic is completely different from using Photoshop’s Unsharp Mask, because Focus Magic works on the whole image and not just on edges. If for some reason I intend to use Photoshop’s Unsharp Mask, I defer that processing until after I have used Neat Image to remove noise and Focus Magic to sharpen the focus.
See this comparison of Focus Magic and Unsharp Mask <
http://www.focusmagic.com/exampleunsharpmask.htm> to see the difference. The difference becomes particularly obvious if you scroll down to Example 3 and compare the highlights in the eyes. Unsharp Mask merely sharpens the out-of-focus halo, while Focus Magic focuses it back down to the circular highlight it should be.
In my opinion, Adobe would be wise to include Focus Magic in their next release of Photoshop. On the other hand, I don’t care if they don’t, because Focus Magic comes out with improved versions much more frequently than Photoshop does.
This thread has dealt primarily with problems that people have been having with the newly released Version 3 of Focus Magic, which was provided free to all current owners of Focus Magic, without giving credit to the very significant performance improvements included in this version, including corrections for motion blur.
— Burton — (not associated with any vendor mentioned)
wow. i may have found my next plugin. after the bugs are worked out of course! 🙂
In reference to the link in Mr. Ogden’s post, I find the example shown of the "sharpening" of the out-of-focus image with the eye-reflection blur to be less than inspiring. Looking at the whole image, there are artifacts produced that – to my eye – make the final image just as awful as the first.
It’s a neat gimmick, and could certainly be of use in a forensic sense, but I’ll pass – particularly given the problems cited in this thread, and the clear buck-passing being offered as "support".
Mr Ogden: Whatever.
The comparsion is only with unsharp mask, it takes a broader knowledge of Photoshop to accomplish the same thing (like median blur with mask). I do photo retouching on a regular basis for large companies, haven’t had a complaint yet. There is a free plugin from Kodak that does much the same thing, I use that occassionally. For the most part I reject crap like that from clients.
ID, do you mean a free trial or a fully functionnal free plug-in?
This problem has been fixed in Version 3.01
3.01? The web page shows V3.0 as the current edition.
