![[Index]](../toc.gif){kind=small}
![[® Previous]](../left.gif){kind=small}
-![[Feedback]](../feedback.gif){kind=small}
|
-
I sent email to S3 for their recent V3.03.16 drivers for my S3 64V (cheap) video card, and were they *ever* pleased!
A copy of the response is here:
This is the second positive feedback I've heard about the new OS/2 drivers. (usually people don't bother with positive feedback) -- thanks! I'll let the OS/2 driver team know.
And the original message is here:
I just downloaded and installed the new OS/2 Trio 64V drivers (V 3.03.16) and they are a true treat...they have changed my generic - and inexpensive - S3 card into a pretty terrific graphics card for my type of use.The installation was simple and straightforward, and the instructions equally well done. Once again - a very big thank you to your developer team and to all others who had to do with these drivers!!
My suggestion for the list of "to-be-thanked" is Matrox Graphics. Their cards are supported extremely well is OS/2 with excellent and up-to-date drivers. I've never had a single problem with my Millennium card in OS/2, and Matrox tech support has answered all my questions in a week, no matter what the OS (I use Win95 & OS/2 4).
Please consider Matrox as the next one.
Perhaps we should send Microsoft E-Mail thanking them for not writing OS/2 applications. After all, if they did write OS/2 programs, many of the great developers who do write OS/2 apps, would not be able to. Microsoft would steal all the ideas and put out half a product, like they have with Windows 95, and Windows NT, and ...!
I'm a Systems Engineer with a major aerospace firm with 25 years of experience developing software. And one of the members of an organization that places the highest possible priority on designing out bugs prior to starting any coding, using the SEI CMM processes.
Chris, it seems to me that you want to use the beta testers to trap bugs that should have been detected prior to coding and long before any beta release. By following good engineering processes, on time, under budget and nearly bug free software can be written consistently. Good design and engineering eliminate the need for your "paying" test force...
I think paying for the job of Beta testing an app is wrong. As a Beta tester, I am providing a service to the developer. I do report in on bugs, usability, etc and expect a two-way dialog with the developer.
Saying that asking for money for a Beta helps weed out those who will not actually test is inaccurate. What you get are those who wish some discount on the purchase price.
Saying that companies can't afford to have a free Beta is wrong. I test for Intuit, Quarterdeck, IBM, all of whom can afford to carry a project. I also have tested for shareware developers. After all, you aren't seeing the app while it is in alpha code. The developer has spent most of the money getting the program to the point where it is fit for testing.
Finally, when I test I have always received (but don't require) a free shrink-wrap copy of the top-of-the-line version of the app as a way of saying "Thanks".
To say that taking the time to look at Beta applicants is too much trouble is nonsense. To say that you can't afford to allow a free test is also silly, as I said, the money has been spent before Beta stage. And to expect people to pay for one of your development stages is plain greed.
After reading the last few feedback sessions, I feel that I have to pitch in on this debate. Two points: first, I am disturbed by Brad Wardell's contention that "A beta program is more than bug testing, beta programs are about ideas and hearing what "wishlists" customers will have before the product is generally available. " If, at the point of beta testing, the developers are still looking for ideas, then the product is far from what I would consider the beta stage. In fact, I don't even know that I'd consider it "alpha" at that point. Any project I've worked on in my career that didn't have a well planned and locked down feature set before coding began has become a nightmare. The implicit statement behind a beta release should be "we think we've tested everything we can, please use this in a production environment and let us know what we've missed."
My second point speaks specifically to the free/open vs paid/closed debate. Another option is free/closed; i.e. those who want to participate in the beta must apply. The developer can then select the participants based on whatever criteria are desired. Testers must agree not to distribute any test software they receive - there are certainly ways this could be enforced. I am currently involved in a test like that for an online service provider and it seems to work well for them and for me.
I agree that we must support the vendors offering OS/2 versions. But I must add that fundementals of good faith are also an issue. Once a consumer has been repeatedly abandoned and shunned, they become leery. I have personaly supported a software company that was entering the OS/2 market, and all too many time I have gotten the short end. I was under the impresion that if a program has potential given support I gave my vote and purchased it. I have several programs that have never worked correctly due to bugs that have never been fixed.
I have come to the conclusion that today's leading software developers have forgotten what is involved in entering a new market. They seem so think that the OS/2 market is an untaped area of revenue just waiting to get taken. I am surprised that they don't realize that, as in any new market, you have to understand what the market wants and develop your marketing attributes. ie: recognition, reputation, technical reliabilty, etc. In summary:
OS/2 is an untapped revenue maker if you invest in OS/2 and do your homework.
We are not a bunch of app-desperate users. Well, at least I'm not. If I need an app that isn't available, I'll make it.
Reconition: If we don't know you're out there or it's too hard to get your software, it will flop.
Reputation: If you are known to bail on your users and not fix your bugs then you will fail.
Technical reliability: I use OS/2 for real reasons. If you are going to put a bunch of Windows programmers in a room and have them make an OS/2 version, don't bother. One reason I personaly left Windows was due to the all too common poor programming. I can accept a bug or two that wasn't found when developing. What I can't stand is an app that will only run half the time on a single computer. Note to programers -- YOU CAN NOT ASSUME!!!. Remember, this is a multitasking OS.
Perceived reliabilty: If you don't develop consumer faith, why should they give you a chance?
Dependabilty: If a customer can't depend on you, don't depend on the customer.
That is an interesting view in putting that much guilt on SPG. I use Warp 3 now and ColorWorks v1 and the usual DOS and 3.1 and am not a MS fan in the least. The way I see it is it looks like IBM is leading everyone into using NT. Did they buy NT? Do they have some NT promotional agreement with MS
IBM sent me their Desktop Software catalog last Spring and it scared me. The cover and first 5-6 pages were an NT promotion using an Astrologers theme. Now after being away for 70 days just this week I have seen an ad by IBM on the Web that has a lightning bolt on a Superman's chest between IBM and NT. "A perfect Team." Take a look at where IBM is leading all the OS/2 users. It is not only us but also the companies and their futures. I am affected by products that fail and companies that bail out but what is by far the worst is when Big Blue starts dropping hints on what lies ahead and it was not a one time publication error.
It's kind of like walking into cold water the first time, you do it slowly. Or to boil a frog you heat the water up slowly or it will jump out.
Amen! and AMEN!! I used WordPerfect for several years and got along fine with its English menus, etc. Then I switched to DeScribe to take advantage of my OS/2, and spent the first 5 hours trying to figure out what all those little pictures meant. And when I wanted to create a couple of customized commands, I had to spend more time in the Icon Editor, trying to make something that was at least understandable.
For an excellent alternative, look at the interface for Golden CommPass. Buttons, but with real words on them. Why not the same thing for word processors, graphics programs, e-mail interfaces (yes, I like PMMail, but I still make mistakes on which button to click).
I just hope that some of the ISV's listen to your comments. At least, please give us the alternative to use words, not pictures.
I have many DOS and Windows programs that fall in the legacy category, but they're not installed. I answered the survey as if you had asked, "how many do you use?" In my case, nearly everything is native OS/2, and I wanted to reflect that.
The last question didn't really have the answer I would have used. I fit the category of 1-3 times per year (booting to DOS).
This month's survey about Java needs more choices. I have installed Java, I have tried Java software, however, I have not been able to get the Java software working properly. Since I don't have another platform to test Java, I don't know if it is my setup, or Java that makes the software not work.
While I have a high end system (Pentium Pro 200, 64 meg of RAM, etc.) I don't like spending time tweaking my setup everytime I install a new software package. If this will be needed with Java apps, I don't see how they will be popular. I realize Java is new, and as such I give it my "Brand New" attention, but even with two weeks of trying to get the softare working, I have not been able to get past a few minutes of use without the Java apps crashing.
Is it just me, or my system, or is this how Java runs on OS/2 systems? I hope this is not the way Java will run in the future. I hope there is someting I can do to get Java working properly, witout having to redo my setup.
Is OS/2's Java engine 100% pure?
As usual, after filling out your reader survey I find myself wishing the choices were labeled differently and that 'none of the above' would be a better match for my opinion.
Do I use Java apps? I have it installed, but must admit that I only 'use' it when I visit a web page with some java-applet. Personally, I don't consider that really using it. When I do run into a web page with java I wish I didn't have it installed. So far these appletts do some mind-boggling trivial thing like some moving graphic. Meanwhile, the poor performance of these programs brings my system down so much I need to reboot. Admittedly, I am behind the times - I haven't installed Merlin yet (I did buy it) so I am using the java that comes with Netscape v2.02. So at the moment, I'd say Java does not enrich my life.
Do I think Java is important? I am an optimist (by definition, I use OS/2 after all) and I think there is an excellent chance that REAL Java apps with REAL performance will be in my future. I may even start trying to write some Java to see what it's all about (however, I would never do much programming in Java except in the unlikely event that it has floating point math capabilities that rival Fortran).
Hey, let's formally invite Intel to Warpstock. Put up the money for a booth & give them free access. They would love it. They are sick to death of waiting for Microsoft to come out with software that even comes close to the quality of their hardware. Mucho good publicity. Whattaya think?
The trouble isn't the cross-platform DEVELOPMENT, it's the cross-platform support. Hiring/training a bunch of support people to handle OS/2 (Assuming you are a Windows ISV) can be a big and expensive hassle. And not worth it, if the number of units is small. I think that, more than anything, is what keeps software from being ported by knowledgeable companies. The lemming companies will never even think about it, of course...
I am a win95 user, not by choice but by demand from my clients who are all using DOS-Win95. I for one would love to be able to use OS/2, and still run my clients WordPerfect Suite 7, or Office Suite 7 on my OS/2 platform. If any one knows a way of doing this, I would uninstall my Win95, and reload my Warp 3 package back on. Anyone out there have an answer for me?
My main OS is OS/2.
And this is where things might seem really weird to everyone. I'm starting a new gaming company (I already have four quality and experienced programmers that are very excited about the products that we will be creating). Our main platform will be Windows.
Why Windows? Several reasons.
Have you every tried to do state of the art programming for OS/2? State of the art meaning BETTER than Quake? IBM couldn't be less interested in helping us find utilities and OpenGL and video cards that have OS/2 support. 3DFX which makes the 3DFX vodoo and voodoo rush chips (and the companies that use that chip on their video cards) tell me to go to IBM for video drivers. IBM, or at least the parts of IBM that I have contacted couldn't be less interested in helping me. Games are not their interest.
I have two choices now. I can either give up totally. Or do something that bothers me morally, write the program for Windows and hope that one day someone will make OpenGL drivers for user affordable video cards that can render complex primitives and textures in 30 fps speeds.
The best of all worlds would be to write this in Java. But Java has even less support than OS/2 for this kind of program.
Any comments or help would be appreciated.
PS: Since I want to be able to "easily" port this to the Mac and Linux too, I've come down to settling on Watcom's compiler.
Just wanted to let you know that there is a new IBM Global Network dialer, although it is still in beta (1.69), it seems to work well so far. This dialer is comparable to the Windows version. Let everyone know! It can be downloaded at ftp://ftp.ibm.net/pub/Advantis/OS2_Client/169beta.exe.
-
Copyright © 1997 - Falcon Networking