ReX

Not even 3.11 boots in 12 seconds (ugh, no thanks). I have a P100 with 3.11 installed so I know for certain too<g>

It's impossible for Win95 to boot from a power on in 5 seconds given the system you describe.

(I assume there's somebody out there who has 95 installed, go ahead please post your boot times).

Nuff said there.

fragg said:

Deep, all the fancy things you and Lut do to troubleshoot, I can't.
I got a lazyman's way. It's called "PartitionMagic 5.0".
(AMD750/256mb/13G Maxtor/GF-32mb/IomegaZIP/HPcdrw/Sony32xCD/LCD).

When I get PC set up, all is running like a charm, I tell PM to make a "mirror copy" of the (3 or 4 gig) Primary Active partition I'm presently in. It does so.
Then I tell it to hide that mirror somewhere on hard drive. It does so.
Then I start messing, experimenting, or whatever in my Primary partition.
If I screw it up -- I wake PM up, and tell it "I'm all screwed up, erase the Primary Active partition I'm in (including all Files, Windows, Everything!) and copy the mirror into my now blank, formatted partition".
It does so. And sets my partition once again, as Primary Active partition.
And I'm back in business, ready and eager for my next total screw-up <g>).
fragg

I didn't describe any fancy things to troubleshoot, so not sure what you mean. I use partition magic for some things, like copying disks, FAT32 conversions and resizing parts.

In actual practice, I can't "mirror" that easily for practical time and size reasons. For example, one machine has 2 30gig drives with 1 partition each. I do "mirror" sections of work that are very important for me. There are several machines with similar configs that are networked. So what we do (if playing around), is go online with a working machine, download drivers, research, etc, a bit different from having just 1 machine one has to rely on.

So different rules according to one's circumstances. What you do is cool for a single machine site. Different hardware can cause problems for one person and yet not for another. So a word to the wise is not to believe everything one reads:)

I hardly ever have the kind of problems alluded to. Last one was a bug in Windows98 update following an IE5 update, but MS showed how to fix it.

Troubleshooting hardware is just something I like to do. Ditto for software stuff. Some issues HAVE to be researched and that's a valuable skill to acquire. I have seen machines that get roasted because users had no idea that certain things they did were just not healthy practices - like skipping scandisk or ignoring that strange noise they hear<g>.

I have no idea why Lutt is doing all these reinstalls, just know that most people don't:))

You like to troubleshoot and mess with hardware and software?
Not surprising. Why?
People like you, Lut, John, etc. have both a natural technical "ability" AND "appetite" to engage in said technical activities.
The 2 go together, like ham n'eggs.
So, I really envy you guys. More power to ya !!!

BTW, the fastest times I ever got my 486-33 w/64mb/Win3.1 to boot to desktop was maybe 12-16 seconds. (Time measured from power-on to desktop).
My Athlon-750 (Win98) boots to desktop in about 30 sec.
My Bro.'s P2-266/128mb/Win95 boots to desktop in about 45 sec.
IMO, Win98 is nicer, easier to handle, than Win95 was. But that's just MO.
fragg

Enjoying messing around with hdwr/sftwr is no different than liking anything else:)) Good thing we're not all the same!

Your times sound about right. All our machines are networked (not the 486), so that takes a tick or 2.

The 486 I mentioned can't run the disk drive at max speed (no DMA possible). This is an OLD machine and I had to run a sofware bios to even access the newer 2gb drive. Ugh. So that's about as bad as it gets:))

Our slowest boot machine has a CDRW using DirectCD. The pwr on boot is a drag because the SCSI card checks to see if it's a bootable device. So all these little things make a difference.

Btw, check to see that DMA is enabled on your Bro's machine (or even a choice) on his drives. This can make a difference in boot speed.

XP is actually faster than I noted - there was a problem with one of the drives (mix of ATA100 & 66). Seems I recall Maxtor mentioning a firmware upgrade I can get. I also think the drivers are not quite up to speed either, but that could just be the debug stuff.

It's all about "MO" - lol.

deepteam said:

Not even 3.11 boots in 12 seconds (ugh, no thanks). I have a P100 with 3.11 installed so I know for certain too<g>

It's impossible for Win95 to boot from a power on in 5 seconds given the system you describe.

(I assume there's somebody out there who has 95 installed, go ahead please post your boot times).

Nuff said there.

No, not from power-on. That stupid BIOS crap takes like 6 seconds. Damn annoying. I had it in work and my boss recommended a BIOS update, he's big on that crap for some reason. Anyways, it moved my BIOS from 3 seconds to 6 seconds booting so I'm trying to find the restore disk but can't find it. As with every other BIOS update we've done at work, it's accomplished nothing.

I was talking about booting through the splash screen, where there is that bar going along the bottom. It's 8 seconds now that I have it hooked up to the cable modem, and another 2 to bring up that silly "Microsoft Networks" login, but without the network setup it made it in about 5. I got a CDR drive now so it takes maybe another second.

The splash screen is NOT a complete boot. Just time what happens after you login (wait for the hourglass to quit).

Lüt said:

I stopped using PartitionMagic 5.0 and 6.0 when I found out they had a lot of trouble with disks over 25GB in size. Basically, once you passed that 25GB marker, it would screw up the partition tables. For example, some of the times I've mirror one hard drive to another and resized the partition to full size, and exit, the partition tables are so screwed up even PartitionMagic cannot read them. Norton Utilities' DiskEdit, however, can. It told me I had a 28GB partition, a 13GB partition, a 116GB partition and a 254GB partition, all on a 25GB drive. The data was there so I just hard to rearrange the partition info with DiskEdit, but it's still a pain and I'm still looking for some new partitioning software as I write this.

Egad, Lut ! Wow!! Never heard of PM screwing up, its considered bulletproof.
But have heard some real negatives lately about Norton.
However, your experience with Norton DiskEdit restores my faith in them.
Is that a standalone pkge or part of N. Utilities?
BTW, I'll check this week with someone at CompUSA, see what the buzz is with his customers re. PartitionMagic on h/drives that are &gt; 24gig. Thanx for that tip.
fragg

It's part of Norton Utilities.

I'm not saying all the Norton stuff is great or top-quality stuff either, it's just good enough to fix the errors we get with PM.

That's all I've used it for so I'm not really in a position to give a product quality report on it, although my boss has had a lot more experience with it and says the Norton Utilities are a generally decent support tool. At least they're a lot better than any utilities Microsoft includes in Windows. The Norton WinDoctor usually catches and fixes a lot of problems with the inner workings of Windows although you have to be careful, sometimes if you have hidden backups of Windows components it discovers those and "fixes" them if you click the "fix all" button after a system scan. For example, you may back up your start menu somewhere and that will find the shortcuts and modify/delete them, and there goes your backup start menu. Of course it gives you an option of what and what not to fix, and detail to go along with it, so just be careful what you choose.

deepteam said:

The splash screen is NOT a complete boot. Just time what happens after you login (wait for the hourglass to quit).

Well, what happens after the splash screen really isn't Windows, it's the additional programs I have in the Run, RunServices and startup. Windows Booting I usually include as from the second it says "Starting Windows" to the time the taskbar is loaded and functional; that's the actual Windows OS coming to.

I totally emptied my Startup, Run and RunServices groups except for 4 system components "LoadPowerProfile, Systray, ScanRegistry, and Taskbar Display Control". I'll start things when I want them, thank you anyways Windows :)

What you "see" is not the same as "what is". Wait till the -hourglass- disappears and you can't "hear" the disk any longer.

However, I just realized the problem you had with Win98 since you mention you are "networked". Win98 can act real bitchy on networked machines "connecting". This problem occurs BEFORE you get to the logon dialog.

So first turn on the "quick" logon, so at least we get that benefit. As I recall, there's also a patch you have to download regards this too. But that still may not fix it:))

Before the patch I had a 5-20 second delay before the damn dialog would come up. So you see, it's not exactly a slow boot issue (nor size, etc), but more a network "bug" issue. You would never see this on a standalone machine.

And then logon -sometimes- had trouble connecting also. I just switched to TCP/IP (since XP does only that), but I haven't consciously tested to see if "all better now".

And to make this even stranger, my machine is the ONLY one that has this strange behavior. The others boot, logon, etc instantly.

As I mentioned earlier, comparing apples to apples is not as easy as it sounds&lt;g&gt; In summary, it's possible for someone to have weird behavior, but that doesn't mean all systems will behave the same way - just "some".

Lüt said:

The bad I found on NU, was it bogged the system, making boot to desktop slower.
I liked it for Registry editing. It lets you Save your orig. registry config in case u screw up. That's a BIG plus.

BTW, I checked w guy at Compusa as I said I would -- he uses PM 5 and 6. Said, no problems have yet occured with partitions &gt; 25gig, but will not categorically say they won't. He'll ask PowerQuest guy next time he's in.
Thanx to ReX, you, Deep and All -- for all the info, see u around the site. Signing off this Thread now.....
fragg

fragg said:

The bad I found on NU, was it bogged the system, making boot to desktop slower.

Oh, you install that crap? Well there's your problem ;)

ONLY run the utilities off the CD when you need them, and nothing else. That's all it's good for :)

I wouldn't install those on my system if they paid me.

fragg said:

You like to troubleshoot and mess with hardware and software?
Not surprising. Why?
People like you, Lut, John, etc. have both a natural technical "ability" AND "appetite" to engage in said technical activities.
The 2 go together, like ham n'eggs.
So, I really envy you guys. More power to ya !!!

BTW, the fastest times I ever got my 486-33 w/64mb/Win3.1 to boot to desktop was maybe 12-16 seconds. (Time measured from power-on to desktop).
My Athlon-750 (Win98) boots to desktop in about 30 sec.
My Bro.'s P2-266/128mb/Win95 boots to desktop in about 45 sec.
IMO, Win98 is nicer, easier to handle, than Win95 was. But that's just MO.
fragg

I really hate computers and anything computer-related.

I happen to know ways around their totally illogical nature but I don't like it anymore than the next person who has to do hours of troubleshooting and researching to perform the simplest of commands which should only takes seconds or minutes.

If anything actually worked, I'd like it. But what are the chances of that? Computers never work, and when they do it's not for long. They really waste so much time it's unbelievable. I just happen to be stuck with them because everything I do involves a computer one way or another. That's why I tried to go into the army after school, I didn't want anything to do with this kind of stuff. Way too frustrating and I just didn't feel like spending all my time outsmarting silly machines - how special can you feel doing that?

deepteam said:

Oh I know all about that stupid hourglass going nuts while nothing is going on on-screen. The thing I like about Windows95 is that it never pulled that crap on me. In 98, I'll finish playing Doom or something, and then exit back to Windows and then it sits there loading absolutely nothing for like 30 seconds. Off my hard drive with you!

The networking problems I know happen because of timeouts on certain hardware and other things, usually crappy networks too. I had those 5-20-second time-out problems when I was trying WindowsME on my computer, but when I went back to 95, the time-outs stopped.

We got a ton of those timeouts running a Windows2000 server at work. Put on NT4sp6a and it ran great.

The other problem I had with 98 specifically on one of my computers was that it didn't support the SCSI card I used for an external CD burner I had. 95 supported it just fine but 98 didn't know what to do with it. Again, drivers and support were useless. Another reason for putting 95 back on. I had gotten a different SCSI card which did support the drive, but that worthless piece of crap went straight to the BIOS, taking an extra 25 seconds before the computer would even hit the Windows splash screen. No thanks ;)

While I'm complaining, APM and APCI power management can get down and suck me. When I press the off button, the machine had better turn off. Simple as that. Then it's on even when it's off? Turn off a computer using that power management and look at the light on the network card - it's still on!

Modern hardware be damned.

Everything worked great back in '96. I wish they could just leave well enough alone and keep making computers that work. With modern software becoming so bloated, it's pointless to even buy a new machine.

The original 95 was extremely unstable (almost like 3.11 - lol) and finally by OSR2+ it settled down. All this just depends on one's machine/software configs. One can't easily generalize all this into "bad" and "good".

Some defaults changed, so always examine what your settings are. Some, like vcache, are hard to find.

New power mgmt + on/off depends on your BIOS and your case. My BIOS let's me use the "front" button to soft power off, but I also have a back switch that turns everything completely off - just like the old days. If you don't have a back switch, get a new case.

'96? Good grief, I remember how pissed I used to get at all the stuff that went wrong then. Todays hardware is not only better, but much cheaper. Especially disk drives, memory, monitors and processors.

New stuff is cool to most of us. The new games would not be possible without the advances in both software and hardware. If one doesn't do any of these things, then old stuff is fine - like a trusty 2400 baud modem:))

My XP now boots in 20 seconds - which is cooking. They just rearranged the way they load stuff - more like mainframes have been doing for 30 years.

deepteam said:

I suggest switching to BSP or other later nodes builders. The one in WadAuthor is the OLD version of BSP that happens to have a flipped side mistake that was only fixed a few years ago in BSP. The symptom is an "invisible barrier" in certain levels.

The newer BSP is also way faster (only an issue for large levels/slower machines).

For the record, I still find WARM to be a godsend when all other
node builders mess my polyobjects up.

No one node builder works in all cases and there will be comments for one or the other - again this area can't be generalized.

However, DeePBSP has options you should change when working with HEXEN type levels with polyobjects. The defaults are mainly for speed. With newer machines it hardly matters anymore. (I think I'll change the default&lt;g&gt;)

Change the depth to 16 - 32 (from 8). Some levels work much better if they are forced into a "split". There are also some alternate methods that may work on some levels. Press F5 and click on the "Node" tab.

You can get an "objective" look at nodes with the DeePsea node viewer. The first screen has a quick summary.

Polyobjects get in trouble when subsectors containing the poly get split. The node viewer can show this problem.

Sometimes you need to move the polyobject just a doom unit over, so DeePBSP has the correct sector. Run DeePBSP with "don't show nodebuild". Then look at the DeePBSP.LOG to see what it found. Compare the polyobject sector number with where they are.

Actually DeePsea should show the same number. It can be incorrect if they are right on a line. Moving it over a unit won't change anything and can fix problems. DeePBSP attemps to avoid splitting around polyobjects, but if the sector is incorrect ....

Hm, looks like this thread finally died :\