Mouse/Cursor Behavior

I'm really having a difficult time with mouse/cursor behavior in Parallels. I was hoping to write Windows software in Parallels, but mouse/cursor behavior is so anomalous that I'm having to work in Boot Camp only.

I'm on a 24-inch iMac with a 750 Gb drive, running a Boot Camp installation in Parallels with 3 Gb RAM. The same mouse issues are experienced with regular Parallels Windows XP clients. None of the issues are experienced when running my XP installation directly in Boot Camp. Here are my issues:

1. Oftentimes the cursor only responds intermittently to mouse movement.

As you move the mouse for instance, the cursor may suddenly be abandoned for so many milliseconds before it resumes moving in synch with the mouse. I've counted as many as 6 interruptions in moving the cursor from the center of the screen to an outer corner.

2. No matter the combination of settings on the Mac and XP side, I cannot get mouse behavior anything like the natural behavior of Windows. Response is erratic and exaggerated.

Is there anything that can be done about this? Behavior is so poor that my productivity suffers so much I can't use Parallels.

Regards,

m

 

Are your tools installed? Our 24" iMacs work beautifully.

 

Tools are installed.

Yes, the tools are installed.

"Beautiful" is easy to mistakenly assume, if you simply go through the installation process and deduce that everything is working as it would normally work under a straight Windows box. But if you are an intensive user, it is difficult for me to believe a person wouldn't quickly notice the great disparity between the behavior of a Windows box and Parallels. Furthermore, as the same disparity appears to exist between Windows in Boot Camp and Windows running in Parallels, I cannot and do not attribute the disparity to hardware.

Other Mac users have been surprised when I showed them that indeed they suffered the same behavior disparity. Typical responses are such that, "Well, I only use the Parallels side to do my flight plans, so maybe that's why I haven't noticed it so much..." and so forth.

As for me, I notice that the typical OSX user operates the mouse with similarly defeated dexterity as under Parallels, and my assumption is that it is possible that OSX anomalies may be affecting the Parallels side. I don't know what good reason defeats people from complaining about these things, but I regularly observe Mac users constantly relegated to performing mouse operations with great care and tediousness compared to the free, productive behavior characteristic to Windows' acceleration algorithms.

Perhaps the best example is Steve Jobs himself, who constantly apologizes during his demos that he is clumsy with the mouse. Curiously, he attributes the clumsiness to himself, while in fact if you carefully test OSX mouse/cursor behavior, you'll find indeed that the Mac often does not invoke a click for instance on the object which the cursor represents to be focused.

That underlying algorithm itself at least needs to be fixed. As to why parallels mouse behavior is so anomalous, I can only speculate.

Nonetheless, I have tried re-installation and several Parallels XP installations to no avail. Intermittent cursor response to mouse movement is a definite issue. But so is the disparity of mouse acceleration under Parallels.

In my opinion, this is really great software -- far too good to be defeated by such poor mouse/cursor behavior.

 

Oh, we have very intensive users, they spend a lot of time on the system every day, with lots of little checkboxes, text fields, etc. If there was a discrepency there is one that I would most certainly be hearing about it from.

Have you tried a re-install of the tools? Enabling/disabling mouse sync? I'm just throwing out common things. If you have, have you submitted a trouble ticket? I have found the support responsive when I have voiced problems.

 

I've written software intensively since 1980, and it has been my uniform experience that it is extremely rare that different systems running the same software truly perform differently.

My box is under heavy stress; but even under light stress I'm seeing the same issues.

If said personnel are experienced Windows users on Windows boxes, I can hardly believe they don't notice any difference, because I can open Photoshop right now and give you just a few operations which will give you fits. Last night I did 7,000 changes in a commercial e-commerce site in half an hour. I started in the Parallels side because I had to open it for a moment. So for the heck of it I tried my work there and I couldn't get anywhere with my changes... so I re-booted into Boot Camp where I could click... click... click... away and get the job done -- largely by deftly operating power tools.

If your users are adamant that their Parallels experience perfectly replicates Windows box behavior as does Boot Camp, I'm frankly incredulous. I bet if I move to the Parallels side right now I don't have to move the cursor from the center of the screen to a corner more than 20 times to get several interruptions.

As for the settings and re-installation, I believe I've tried everything I could discover. But I suppose I can give it a go again and report back if anything is fixed. I've tweaked everything available, including combinations of Mac and Parallels settings.

As for the trouble ticket, I will try to re-exhaust the potential configuration issues once again before I do so; but frankly, I'm convinced there's an underlying software issue.

 

Do they spend half their time in Boot Camp, and do they notice no difference?

 

I'm going to wait now for a reply to my issue report.

 

I guess we were using the word 'intensly' differently. I meant as in using the computer all day, rather than driving the processor hard. I do not doubt, nor did I doubt, that you had experience with computers. I only started with Parallels tools and basic things as just about all the questions I see are answered by basic things.

I just opened up my Parallels and ran in Full Screen and Coherence (I am running a 15" MBP, so perhaps that makes things differ) and moved my mouse around probably a hundred or so times with absolutely no interruption. I have found the click to occur on the very tip of the mouse pointer.

I don't mean to say any of this from a stand-point of argument, just speaking my experiences and throwing them on the table.

Perhaps you can take a screenshot video and post it?

I would recommend getting that trouble ticket opened up whenever you feel ready. It sounds as if you sent one in. Hope things get worked out.

 

I work 15 hours a day, and most of the time 7 days a week. By intensely however, I mean both 1) stressing the limits of software by regularly exercising its extreme capabilities, and 2) necessarily running multiple applications which may for instance stress thread support as might affect input device support.

Opening up a single or even a few applications may never replicate the conditions (even as I experience them even under light load). For valid tests, open email, your browser, a web development application, Windows Explorer, and Skype. I have all that running in Boot Camp right now, plus 5 open emails I'm responding to -- just regular usage; and I promise you I'm having the troubles I report in Parallels under these usual loads.

As to where you get the idea a person could even *possibly* take a screen shot from which anyone could detect interrupted cursor response I don't know.

As to a video of screen exercises, that requires special hardware. According to the documentation for Windows Movie Maker for instance:

"You can capture video and audio to your computer by using Windows Movie Maker. To begin capturing, a video capture device must be connected properly and detected on your computer by Windows Movie Maker. Some audio and video capture devices and sources that you can use include a digital video (DV) or analog camera or VCR, a Web camera, a TV tuner card, or a microphone. You can capture content live or from video tape."

Regarding the pointer anomalies of OSX itself, your response leaves some doubt that's even what you tested.

 

Mike,
I'm finding your posts a little, uh, annoying to wade through. Sorry for trying to help.

Frankly, it's great you did 7,000 changes to an eCommerce site in an hour. But when it gets down to it, do we on the forum really care? No. Unless you are trying to market for more work. You work 15 hours a day, mostly 7 a week, great. You are hard-working. But does that really pertain to the issue? Once again, no. That you have 5 e-mails you are responding to and all those other applications doesn't make me think you are 'that much better than the rest of us weaklings'. Maybe you do more and much better work than I do, but is that really relavent? Do I really care? I was trying to help you diagnose a problem.

As to the idea of the screenshot video, if you used OS X to record the screen and the cursor was interrupted in the movement, you could see it, no? There would be a difference in acceleration or it would interrupt as you say it would. It's actually not a question, but a fact.

If it is an OS X issue as you say at the bottom, why the heck did you bother coming to Parallels about it? Did you want them to patch OS X?

Long story short, sorry I bothered trying to help you. It sounds like you don't want the help as you already have it figured out farther than any of us could probably 'dream' of taking it. Why? You have the better mind than all of us as you leave reamaining to be said by use of an enthymeme. Whether or not you meant it, you did.

 

The fact is Eru, you haven't even answered my question whether the "intensive" work you cite even involves the obligatory comparison to behavior in Boot Camp. Quite obviously, your post meant merely to dismiss my report offhand by the dubious proposition your intensive co-workers' skills and experiences somehow overrule mine. Exactly how, you do not say -- thus the reply. Simple matters of use; dismissal of an issue by mere reference to ostensible "intensive" use not even necessarily qualified by bona-fide comparison to XP box behavior, and still you're saying everywhere I'm just a fool. The point is, your irritation results from dismissing my experience by ostensibly *more* intensive use on your end.

As to your further unqualified assumption that I should instead have complained to Apple ("If it is an OS X issue as you say at the bottom, why the heck did you bother coming to Parallels about it? Did you want them to patch OS X?"), well now, I do assume you are the moderator of this forum, and that your answers are to represent Parallels.

So once again, am I wrong to interpret this as anything but further stonewalling? You first dismiss that the reported problem exists -- saying your intensive use (where cursors are at least moved around the screen) *shows* no such behavior (which I really doubt); then you expect me to go elsewhere to get it fixed. That's your idea of support?

As you should know, Apple is only bound to die laughing when I come to them with my report of Parallels behavior, wanting Apple to rectify *something*. As you should further know, whenever you design an application to run on top of any other technology, the job is to make your application run *right* on top of that technology. That means from the start that you are taking it upon yourself to provide mouse and cursor behavior without the described anomalies. A lot of people will call your behavior satisfactory; but every time I mouse toward something in a long series of moves to get a job accomplished in due (customary) time, and the cursor stops not even just once, but several times in the mouse move, a *huge* impediment is imposed on my progress. I don't just object to that; I have already tried to tell you it makes it impossible to use your application. A friend of mine with a mess of Macs is glad they didn't try Parallels yet, seeing my mouse behavior. I can tell you what an irritation is if you want.

You don't seem to have anything to offer but dismissal, and yet you call that help.

As to your assertion, "As to the idea of the screenshot video, if you used OS X to record the screen and the cursor was interrupted in the movement, you could see it, no? There would be a difference in acceleration or it would interrupt as you say it would. It's actually not a question, but a fact."... there is no such fact.

How even would you know such a thing? Have you actually recorded a video of XP cursor behavior from OSX? Is there a process for such a thing?

If what you have said were true, then in giving the Mac side control so that I could record the video, the Mac side is going to be recording behavior in the guest operating system.

Just exactly how would that be? (Or is this just a further runaround to which you expect no response -- that I just go quietly with my non-existent issue?)

Here's what I think is going on: Developers who may very well be entirely unfamiliar with Windows cursor and mouse behavior gave themselves a little hooray that the cursor moved with the mouse -- and they simply accepted the anomalous behavior as if, despite its substantially preclusive/objectionable performance, this were just ordinary Windows box behavior, or sufficiently satisfactory Windows box behavior.

But it's not.

 

PS.

I will look into shooting a video from the Mac side, just in case it is possible a process will record behavior in a Parallels window.

 

I thought I was the one who pointed out the double use of terms after you questioned it.

Sorry I couldn't be more clear. I never said we had more intense use, either. I pointed out a different use of the word.
Secondly, 'the obligatory comparison to BootCamp', does that mean I see how the mouse works there? In what manner? The speed of the mouse? Acceleration based upon motion? The fact that it can move across the screen uninterrupted? This is probably my bad in missing something obvious.

I'm not the moderator of the forum. I was raising the question. You point out that I did not test the irregularity of OS X cursor control, if my failing to do this is relevant to the problem, then it is at least part of the issue, right? If it is not part of the issue, then why is it relevant?

See above. Maybe I'm the one being daft.

I said:

Once again, perhaps my comprehension is so sadly behind yours I cannot follow your implications.
I have provided no help as we have got no further on the issue. I was trying to narrow down the issue as to whether or not it was a glitch across the board. I saw the difference in load once I realized the double use of a term, but you said under 'light' useage the matter does not occur. I thought we are having a disconnect in me understanding your problem, and so, I asked for a video capture. Is not that reasonable?

Then what is the problem? The cursor is graphical, and thus the motion of its progression from coordinate x1,y1 to x2,y2 can be recorded. If you demonstrated movement in a set method and it's lack of moving appropriately you would have a point, right?

Well, the one that comes first to mind is to mirror the display with a display out, drop that to a DV bridge connected to another computer and then record the video there. I also know that Snapz Pro records the screen and will save the file.

Just to sum up the personal side of things, sorry we are miscommunicating, I'm trying to find the disconnect. One of the methods of troubleshooting I use is to isolate the issue by eliminating possibilities. I see that I was/am communicating this in the wrong fashion. Furthermore, while I pointed at you talking about you detailing all your qualifications being not relevant, I did not intend to do it to shoot you down. While a little background is useful, throwing numbers of your success while people are having a disconnect is just highly frusterating to the other party involved in the discussion.

Sorry to make this painfully long.

 

Anything here that would solve the issue? Where's the beginning and ending summary?

 

It doesn't appear you provide a process for your assertion that works by OSX processes. I further doubt you have ascertained you can perform said process with the additional gear you describe.

HAVE YOU DONE SO?