PC still playing Amiga catchup

[Karlos]

I'll post the A1 boot time when I have a moment, right now I need to get ready for work. I suspect it'll be the fastest of all of them.

[Karlos]

Think base frequency, not QDR or DDR, your FSB has a base frequency of 333 (4 x 333 = ~1333) and your RAM has a base frequency of 667 (2 x 667 = ~1333), at the moment you are running 1:2, try underclocking the RAM to a base frequency of 333 and then drop your latencies.

-edit-

Scratch that, it's already running at 333x8 (dual channel interleaved). DDR3 latencies, eh?

[Karlos]

You have a Frankenstein A1200.

No, I have an expanded A1200. Nothing frankenstein about it. The PPC card is a standard trapdoor expansion. I'm using the basic IDE port.

My system: A1200, 68060, 32 meg ram, OS 3.1 (Magic menu, tools menu, magicwb, toolmanager docks, 1 gig flash card):  

Boot up time when hard drive light stops and i can use the OS without delays, stutters :4.0 seconds.

 


You are using OS3.1. Try upgrading to OS3.9. Frankly I'm shocked you haven't. Call yourself an Amiga user?

Besides, my boot times were from switching the machine on from cold. Any sucker can quote their boot time from the moment the OS starts booting. If I do that after a fresh reboot from RAD I can get it down to about 6 seconds, on the 040 which is ~1/3 the speed of the 060.

Shut down: flick of a switch ( how fast can you do that)

Good luck beating that.

The 1200 and A1 can be switched off like that. So can the PC, as long as you don't mind corrupting your data, which is equally possible on the Amiga if disk writes are happening :rolleyes:

[Karlos]

The PPC card is NOT standard anything.  It is a rare add-on, that was intended to act as bridge in the Amiga's move away from 68K to PPC. The owner of Amiga had no input or control over it.There are major performance bottlenecks that occur in that card due to to context switches, especially when you run 68 k software, the PPC has no caches etc, the memory busses are designed to work with a 68K CPU in mind.  Stick to Classic Amiga ie 68K for a valid comparison.

In case you hadn't noticed, the PPC board has a 68K processor that runs pretty much everything in OS3.x. The PPC itself is totally irrelevant as regards the boot time.

 

So are mine.

You said "from the moment the hard disk light goes out".

Not EQUALLY POSSIBLE especially if you use PFS 3 or SFS or, if you use FFS, ... *snip*

There's no swap file because you don't get any virtual memory support in 3.x. One other basic feature taken for granted on pretty much every other modern OS.

[Karlos]

Eh? :lol:

Try running the CPU at 7 x 400 (QDR1600), that'll put you at 2.8GHz, you shouldn't need any voltage adjustments, then drop your RAM to 400MHz (DDR800) and theoretically you shouldn't have any troubles running your RAM latencies at 4-4-4-12. Or then again, you could just leave it all stock

From wikipedia

Latencies

While the typical latencies for a JEDEC DDR2 device were 5-5-5-15, the standard latencies for the JEDEC DDR3 devices are 7-7-7-20 for DDR3-1066 and 7-7-7-24 for DDR3-1333.

DDR3 latencies are numerically higher because the clock cycles by which they are measured are shorter; the actual time interval is similar to DDR2 latencies (around 10 ns). There is some improvement because DDR3 generally uses more recent manufacturing processes, but this is not directly caused by the change to DDR3.

As with earlier memory generations, faster DDR3 memory became available after the release of the initial versions. DDR3-2000 memory with 9-9-9-28 latency (9 ns) was available in time to coincide with the Intel Core i7 release.[7] CAS latency of 9 at 1000 MHz (DDR3-2000) is 9 ns, while CAS latency of 7 at 667 MHz (DDR3-1333) is 10.5 ns.

That makes 6-6-6-21 (the current settings) look pretty good from here...

[Karlos]

Is it? I seem to recall that the 68040 on a ppc card was slower at the same clock speed than a non-ppc 68040.

Only when the PPC is accessing memory and the 68K wants to. Otherwise not really. The PPC cache is ample enough that the critical bits of the warpos kernel can fit in there whilst it's doing nothing.

[Karlos]

Yah, 6-6-6-21 is nothing to be sneezed at, now overclock that sucker!!!

When the CPU is cheaper to replace, I'll do it :lol:

[Karlos]

Chicken...buck-buck-buckaw! :lol:

Stock intel CPU cooler, you see. Having said that, it isn't bad. I haven't seen it go over 45C yet.

[Karlos]

A few people here in this thread have referred to items that are NEW or nonexistent (exist in spec only).  Those are useless for a common application.  I have written accurate timer using 8253 under Windows XP, but it's accuracy is 840ns (less than an Amiga).

Joyports are nonexistent on PC's for some time now, but you are insisting on using it in your argument.

Amiga's accuracy is 558ns and HPET cannot be used to compare with it and RDTSC has problems as already mentioned.  I just remembered another problem with RDTSC-- it's nonserializing; so I also have to make sure I write some special instructions after and/or before it to prevent it from executing in parallel.  So much for your speculation that it's accurate to a few machine cycles

I think you'll find I already pointed out that as a requirement when writing the loop, along with locking it one core on multicore and preferring constant TSC where available, or preventing speed step from altering the core speed. Not my fault if you can't be arsed to read the CPU documentation properly.

Also, other than dismissing it outright, you haven't given any reason why HPET is unsuitable for your timing requirements.

>Re: bootup times:

>From cold, timed this morning:

I think one good test for boot-up time is is how long it takes to boot-up until you can run your specific program that uses OS functions.  You can launch Amiga programs directly from command line via startup-sequence.

No, that's an entirely synthetic test invented by you. Under your stated circumstances, my chosen application is Workbench...

[Karlos]

Did you miss a post regarding RDTSC from me?  I stated that I shut down power management to prevent RDTSC from misbehaving on a Toshiba and the system overheated.

LOL, excellent

How do you prevent processor from going  into SMIs and C-states and stopping tick counts?  It's not in any Intel reference I have.

Did the part not actually have a constant TSC register? Most current intel processors do.

>Also, other than dismissing it outright, you haven't given any reason why HPET is unsuitable for your timing requirements.

The fact that it's not in any of my 10 PC systems is good reason to dismiss it for now.

Are you sure? It could just be that whichever version of the OS you are using doesn't let you use it.

 The fact that it won't work on majority of systems out there is another good reason.  The fact that it's accuracy is affected by other IRQs in the system is another good reason.  I can mention more but why waste time with it if I can't use it generically.

But this is true on the PC for most things. Very few applications that need to bash the hardware directly are going to be universally compatible

By the way, Amiga timer is 558ns w/o +/- bullcrap (as I stated); HPET has a +/- latency as well attached to it.

Maybe not, but you can't possibly be suggesting that whatever code you execute every 558ns isn't going to add latency.

>No, that's an entirely synthetic test invented by you. Under your stated circumstances, my chosen application is Workbench...

If you use Amiga for some specific application, it can launch faster before you load WB.

Well, by the same token, I could roll a linux kernel that gives me just what I need to run "nano" and edit a few files.

The point is, I haven't used custom OS boot disks on an amiga since the day I got my first hard disk. Which was a long, long time ago.

[Karlos]

It only talks about fixed counter using bus clock as opposed to processor frequency, but even the bus clocks vary with systems.

Given that you can query this property, where is the problem?

[Karlos]

Give any sane person just the two choices below:

1) Fast boot time

or

2) Little need to (re)boot once up

Guess what they pick?

Being able to boot faster might be important to casual users, but for those of us that actually do real work on a PC, system stability once booted is absolutely paramount. Nothing else, whatsoever, is anywhere near as important.

My Amigas all crash, not too frequently but usually without any warning and at the most inopportune times. They can't stay up for extended periods. The (classic) OS suffers memory fragmentation, even with patches to help minimise it. AmigaOS might boot fast, but it isn't a safe working environment. I've lost count of the number of times I've lost work on it. Now I use KDevelop and edit source code for amiga projects (yep, I'm slowly getting into it again) in a directory on my linux machine that's shared over the network with the A1 and A1200. This way, I never lose anything. Why? Because Linux is far more stable. You don't get processes trashing each other's memory. You don't get large allocations splitting the available ram into two non conttiguous  blocks, you don't get sudden deadlocks caused by magic menu and StormC's built in editor failing to get along. You don't get the results of your own accidental bugs in assembler code you didn't notice at 2am bringing down the whole machine in technicolour glory...

[Karlos]

Hibernate works nicely here.

In the few times I've bothered to hibernate Vista (since I don't tend to use it often), it's woken back up again to normal operation in less than 30 seconds.

[Karlos]

@ Karlos
@meega.

I see.  The solution to slow boot times is ........don't turn of fthe computer. May be OK for the minority-yes the majority of users power their machines down.  My A1200 hasn't crashed in many months: but then I don't have a Frankenstein hack, I don't use software written by amateurs.

Try running your amiga for several weeks continuously and see how stable it isn't.

When you say slow, you are still talking about a minute at most. Which is only slow if you suffer from some form of attention deficit disorder. If your boot is taking longer than that, perhaps you have a totally clapped out unmaintained heap of scrap for a PC, because it really shouldn't.

So you admit the PC is still playing catchup with the Amiga then.  No amount of work around will change the fact that PC's take a factor of at least 10x longer to boot, despite having clock speeds in every hardware component that are factor of 1000 faster.

Nope. See my earlier post. My PC boots faster than my A1200 running an otherwise base OS3.9. Whereas my A1200 loads a fairly minimal OS3.9 for my needs, my PC loads a full desktop OS and an entire suite of server services. It may be a factor 100x (not 1000) faster by CPU clockspeed, but it does more than 100x the amount of work, every second it is up.

[Karlos]

On vista on my laptop thats a few months old, I can resume fairly quickly from sleep, but resume from hibernation is not apprecaiably faster than cold botting: why would it be.  hibernation just saves everything to the hard drive and opens them up again when you resume by reloading from hard drive.

Hmmm, what you want there is a solid state drive :-)

[Karlos]

My A4000 6-7 years ago was a used as a server for 6 months straight without rebooting.

Running a basic OS3 or OS3.1 install on the base machine, possibly. Even then, you were lucky.

One minute?  We've been down that path with your souped-up, hardware hacked/overclocked, custom_OS running PC that you run 24/7.  You are in the infinitesimal minority of PC users.  Your experience doesn't count.

:roflmao:

Put your toys back in your pram and think about what you just wrote. My experience doesn't count simply because it is contrary to yours? A PC made entirely of off-the-shelf components, only one of which is overclocked (the graphics card) and running the most popular open source OS in the world counts as a hacked/overclocked custom OS PC? Do you have any idea how ridiculously petulant you sound?

I can name two other users right here on this very board that have PC's in the same hardware class as mine. 2GHz multicore is not a minority, it is pretty much the norm for current generation machines.

Further you've resorted to calling me and the millions of Windows users and MS itself, for whom boot times matter, "insane", and we are all "suffering from attention-deficit disorder". Name-calling is the last resort of those who simply can no longer defend their argument.

No, I've suggested that if having to wait 1 minute is too stressful for you then you probably have it. Not the millions of others for whom it is a minor inconvenience at best.

We've talked about your frankenstein A1200 before.  It doesn't count.

Bit hypocritical of you to slate my A1200 for having 2 CPU's and as much memory as I could throw into it. Why isn't it "souped up", like the PC instead? It's expanded with entirely legitimate expansions. There's nothing sat hanging upside down from one of the custom chips or dangling off the clockport. There aren't any low level hacks running in the OS.

By the sound of it, your definition of frankenstein is any amiga with a faster CPU fitted.

You have 4 CPU's running at 2,400 mhz. 4x2400 about 10,000.

One 14 mhz 020 boots faster than your 10,000 mhz of total CPU power machine.'  Nuff said.

Of course, if you knew the first thing about multicore computing you'd know that performance is not a simple linear scale up of clockspeed x cores. Also you'd know that only threaded applications or multiple concurrent instances of single threaded applications benefit.