A Blast from the Past! Part 2

[Tom Allensworth 1,532]

The second editorial published in our first edition (downloadable version):

----------------------------------------------------------------------

 

 

 

http://' target="_blank">

 

Pushing the User Envelope

 

We all know what will happen when you get too close to your aircraft's performance envelope. Unpredictable and sometimes fatal things can happen. Exceed that envelope, and sure as you are reading this, what was a possibility becomes a certainty. You will die (in simulation of course).

 

Once, not too many years ago, purchasing software was rather an easy undertaking. You went to your nearest software retail outlet and purchased the title of interest. You were reasonably assured that it had been stringently tested and that what you were purchasing was sure to work on your old and trusty DOS machine - in short, you were reasonably assured that it would stay well within the performance envelope of your system. You could be assured that the program you purchased would at least load on your machine, and work in some fashion, if not exactly as the manufacturer said it would.

 

That has changed considerably in the last 18 months or so. Since the advent of Windows 95, and the proliferation of add-ons and "tweaks" to improve its performance, your successful installation and use of a program can be a hit or miss afair which gets very close to the edge or exceeds it with stunning and fatal results. With the unleashing of Windows 95 (and some would argue much sooner than that) the demands that gaming software houses are putting on our hardware, underpinning operating systems and our periphreals are in a lot of ways, exceeding the ability of the user to install, much less, use the software. Fatal encounters with the edge.

 

Take as an example the recent release of U.S. Naval Fighters 97 by Electronic Arts under the label of Jane's Combat Simulations. We purchased a copy of this program not too many weeks ago and found out how true the above really is. What may in fact be a great product, for us, was never realized. Requiring the installation of DirectX 3.0 components, USNF97 takes the three most sensitive underlying components in a system (video, sound and interface drivers) and rehashes them to meet the requirements for DirectX 3.0 certification - certification which must be attained BEFORE USNF97 will work. We wondered why EA would choose to use version 3.0 when Microsoft doesn't even do so in its FSFW95? Is this approaching the edge?

 

Following EA's instructions to the letter, we installed USNF97 and with excited anticipation rebooted the machine. Ran the game and NADA, Zip, nothing. Hung the system more securely than any other game of this type that we had encountered. Checked the certifications and all were set to the required 3.0. Repeated the process two or three times (again following EA's instructions on Un-installing before Re-installing to the letter). Never did get beyond the opening screen. We decided to take a break and go flying with Microsoft's FSFW95. FSFW95 had worked reliably and without too many hiccups. Now that wouldn't work. Seems whatever EA's USNF97 did to the DirectX drivers did FSFW97 a great disservice in the process.

 

To make a long story short, we ended up reinstalling all of our latest Stealth 64 drivers and the Diamond Media control panel (why in the good lord's name did we have to do that?) and finally got FSFW95 to work again. We ended up writing EA's online support folks explaining the problem and two weeks later (TWO WEEKS!!!) got back a response. Paraphrasing, the response went something like this:

 

"Sorry for the delay in getting back to you... Please remove your Diamond Stealth control panel and reduce your screen resolution to 640X480. Oh, and make sure all of your Stealth 64 drivers are the most recent releases and that your Creative CD's are as well."

 

Say what? You just sold us a program that pushes the envelope and now you want us to tweak it back to the origins of time to get it to work? Hmm... Okay, so we will play along. Did exactly as they suggested, and guess what. Nothing. No change. Still Hangs. Have not ever been able to get it to work. We now have a great drink coaster and a manual that is added to our library of "Would have loved it, but... " software.

 

Conclusion:

 

This is not a slam at Electronic Arts by any means. We have had our share of problems with virtually every software manufacturer out there. Not too many issues hit the street in a flawless state (come to think of it, we can't remember one title that got it totally right the first time). What it is.... This is a scream for rationality in the gaming development and publishing world.

 

Manufacturers, like EA, will argue that we want it both ways. We want the latest, greatest, most up to date, whiz bang simulation that we can get our hands on. They will argue that in order to compete they have to push the edge. They will argue, convincingly, that in order to meet our demands, they must challenge the technology. Yeah, we suppose that we do want our cake and be able to eat it too. But....

 

We need some rationality. Here are our suggestions to industry:

• Aim for the lowest common denominator (LCD) of user configurations that will also meet your design demands for performance and enhancements, but then test and test again to insure that the title works flawlessly within lowest common denominator range
• Beta Test across as wide a range of user demographics as you can find that meet those lowest common denominators and then beta test again.
• Use proven components in public releases (publishers must beta test when pushing the edge - but why in the good lord's name use a component that the author of the DLL is not even using - and to make matters worse, is rumored to be jumping one revision or more up to get away from flaws already present?)
• Cover the bases on interfaces - aim for interoperabiliity in at least the 70 percentile range of user configurations and 100 percent within your LCD goals
• Push the envelope using a community of users who are prepared to test and test again, and who recognize that Beta Software can and will blow up, often taking other components with them.
• If you are going to push the edge in final releases, at least get your user support system right! And for goodness sake, don't brag about being a FREE support system. Your's is not free. We paid good money for your product. The least you can do is make it right without charging us for the privelage.

 

Electronics Arts is not a shlock house. Its a great company that aims to please and do it right. They have historically published excellent products and have provided their customers millions of hours of enjoyment and pleasure through their produced titles. EA just happened to be the most recent of our software evaluation experiences and one that really got us fired up about the situation today with gaming technology and the pathetic state of support mechanism available to consumers. We know that there are probably many of you that have purchased USNF97 and don't have a clue as to what we are talking about - your installation went just fine, thank you very much, and have never ever suffered a problem or had to deal with the EA support system. A roll of the dice? Absolutely. But that is exactly the point of this diatribe. When it is a roll of the dice as to who will or will not suffer from exceeding the performance evnvelope, then the fundamental principle of software design should be to attain the most reasonable of common denominators - not the lowest, but at least those that your packaging side panel says that you do attain.