Microsoft certainly knows how to write an optimal caching algorithm
I take it you've never compared SMARTDRV with PC-Kwik, NCache, CPCache, or just about anything else that was competing with it?
What needs to be made clear is that they are using an approach which effectively disables the ability of Windows 95 to multitask
Source please??? I really do want to read this somewhere besides the SYCR thread.
I grabbed the following from an SYCR news release: The second is
Multitasking Acceleration which significantly reduces the performance degradation associated with Windows 95's multitasking of the hard drive
This could be taken many ways, but I don't read "disables multitasking" in here anywhere. Might it be true that it disables multitasking until it's needed? Many programs for DESQView did exactly this (preventing applications from being polled unless the application itself were requesting the CPU's attention) and, speaking from experience, they *dramatically* improved DESQView's overall performance, both when multitasking was needed and when it wasn't. TAME was the name of the most popular one.
My guess is reviewers will expose vast numbers of problems with the product and it will not even get off the launch pad
I'll be eagerly awaiting reviews. Whether it will leave the launch pad is no more than a guess at this point.
There may even be a problem getting Windows 95 compliance certification
Again, I'd like to see the source of this information.
I just spoke to John Shaw at Syncronys and have to admit I didn't come away from that conversation feeling warm and fuzzy. His answer to both of my questions (disabling multitasking, Windoze95 compliance) was "We haven't released the product yet." I thought he was being needlessly coy on the subjects, however, in fairness to them, it *is* true they haven't released the product. I'm sure it's out in beta, but most beta agreements include a non-disclosure clause.
If I can't get information to the contrary, I'm wondering where you're getting your information.
Worst of all, use of the product may void the warranty of computers that use it.
Again, source please?!? How on earth can use of a software product void a hardware warranty??? I really have to wonder about your motivation when you throw out a statement like that. That statement seems completely unfounded and very irresponsible in a thread read by so many investors, most of whom lack the in-depth knowledge necessary to decide for themselves the veracity of such a statement.
I haven't encountered many ways for software to damage hardware. I've only been able to do it once, personally, and I had to do it very intentionally. Specifically, I used Assembler to endlessly attempt to drive a hard drive's read/write heads into the spindle. The drive eventually failed, but inspection of it revealed I never successfully wore down the heads like I was trying to. Instead, I trashed the stepper motor. I repeat, the only way to do that particular trick is to try very specifically to do just that. It's not something a "bug" can cause to happen.
Please list your sources. |
