home *** CD-ROM | disk | FTP | other *** search
- Received: from nacm.com by cyber1.servtech.com; Sat, 4 Nov 95 23:50 EST
- Received: (from majordom@localhost) by nacm.com (8.6.10/8.6.9) id SAA14678 for executor-digest-outgoing; Sat, 4 Nov 1995 18:47:15 -0800
- Date: Sat, 4 Nov 1995 18:47:15 -0800
- Message-Id: <199511050247.SAA14678@nacm.com>
- From: owner-executor-digest@nacm.com
- To: executor-digest@nacm.com
- Subject: executor-digest V1 #324
- Reply-To: executor@nacm.com
- Errors-To: owner-executor-digest@nacm.com
- Content-Length: 7044
- Content-Type: text
- Precedence: bulk
-
-
- executor-digest Saturday, 4 November 1995 Volume 01 : Number 324
-
-
- In this issue:
-
- Re: Pentium Pro (i.e. P6) introduced today
- SoftWindows someday?
- [none]
- E/D vs E/L speed
- Re: E/D vs E/L speed
- totally unrelated question
-
- ----------------------------------------------------------------------
-
- From: dan_g@ix.netcom.com (Dan Guisinger)
- Date: Fri, 3 Nov 1995 20:26:17 -0800
- Subject: Re: Pentium Pro (i.e. P6) introduced today
-
- You wrote:
-
- >
- >References: <m0tAvio-00000EC@beaut.ardi.com>
- > <Pine.LNX.3.91.951103165025.603A-100000@spam.blenke.com>
- >
- >>>>>> "Ian" == Ian C Blenke <iblenke@blenke.com> writes:
- >
- > Ian> On Thu, 2 Nov 1995, Clifford T. Matthews wrote:
- > >> I do not want to start a processor war on this list, so I will
- > >> not give my opinion on x86 vs. PPC vs. Alpha vs ...
- > >>
- > >> However, if anyone can get one and run Speedometer under
- > >> Executor on one, I'd like to know the numbers that it gets.
- > >> Speedometer 4.0 numbers would be useful, too, although you'll
- > >> have to find a copy of Speedometer 4.0 and some of the numbers
- > >> will be (obviously) incorrect.
- >
- > Ian> I'm a bit confused here... there exists both PPC and Alpha
- > Ian> ports of executor? I know that it is entirely possible (your
- > Ian> 68040 emulation is downright scary), however from what I
- > Ian> understand you still need to port some instructions a bit for
- > Ian> "native" host processor speed.
- >
- >Poorly worded. The subject line is: "Pentium Pro (i.e. P6) introduced
- >today" and when I first wrote the post I asked if anyone could get
- >one, to please send me speedometer numbers, then I thought about it
- >and decided to add the first paragraph in an attempt to stave off
- >architecture wars. Sorry about the confusion.
- >
- >BTW, we did have a version of Executor running on the Alpha for a
- >little while. Expect to see one again, some day.
- >
- > Ian> Granted, you could run Executor under a SoftPC emulator (one
- > Ian> would suppose), but the mere thought makes me a tad ill ;)
- > Ian> (SoftPC emulation under Executor, under a Dosemu session
- > Ian> under Linux...)
- >
- > Ian> *cough*
- >
- > Ian> - Ian Blenke <ian@blenke.com>
- >
- >--Cliff
- >
- >
-
- Well, on Intels Pentium Pro area, they have testamonies from people who
- were lent those machines, and they still should have them for a couple
- months, so, see if you can get their names and e-mail addresses, and
- contact them!
-
- - -=Dan Guisinger=-
-
-
- ------------------------------
-
- From: billc@pegasus.attmail.com (WJCarpenter)
- Date: Sat, 04 Nov 1995 15:52:23 +0000
- Subject: SoftWindows someday?
-
- The recent review of Executor in the Nov95 _Linux Journal_ showed the
- reviewers running SoftPC. Is there any chance that SoftWindows will
- someday run under Executor? (If you're a Linux user, you probably
- know why I ask.) Speed of execution is not necessarily an issue as
- far as I'm concerned.
- - --
- Bill@attmail.com billc@pegasus.ATT.COM or
- (Voice/Fax) +1.908.576.2932 William_J_Carpenter@ATT.COM
- AT&T Bell Labs / AT&T WorldNet Services LZ 1C-320
-
- ------------------------------
-
- From: William Yeh <talon@interport.net>
- Date: Sat, 4 Nov 1995 13:16:37 -0500
- Subject: [none]
-
- I tried running Bolo and the following error occured:
-
- General Protection Fault at eip=00ac459d
- eax=003015da ebx=90abcdef ecx=001dc760 edx=90abcdef esi=efcdab90 edi=90abcdef
- ebp=00194dd0 esp=001d8798 cs=00af ds=00b7 es=00b7 fs=00e7 gs=00d7 ss=00b7
- Call frame traceback EIPs:
- 0x00ac459d
-
- Does anyone know what's wrong?
- -/- Talon -/- talon@interport.net -/- Alt -/-asics@cyberspace.org -/-
- - -/- proteus@nether.net -/-
-
-
- ------------------------------
-
- From: brianw@scripps.edu (Brian Wimberly)
- Date: Sat, 4 Nov 1995 11:56:55 -0800
- Subject: E/D vs E/L speed
-
- Which is faster on the same hardware -- Ex. /DOS or
- Ex. /Linux?
-
- I recently started using Linux and I am wondering if
- Executor would be faster on my medium-to-low-end
- hardware (486 DX2 PCI, Trident TGUI9440AGi garphics card).
- I imagine the speed of the Linux version will improve
- once my graphics card is fully supported by XFree86.
-
- Brian
-
- ------------------------------
-
- From: mat@ardi.com (Mat Hostetter)
- Date: Sat, 4 Nov 95 13:20 MST
- Subject: Re: E/D vs E/L speed
-
- >>>>> "Brian" == Brian Wimberly <brianw@scripps.edu> writes:
-
- Brian> Which is faster on the same hardware -- Ex. /DOS or
- Brian> Ex. /Linux?
-
- Our CPU emulator will run at the same speed under any OS, so there are
- only a few issues:
-
- 1) How fast can Executor access the graphics hardware?
-
- 2) How efficiently is virtual memory implemented? (Only relevant if
- you don't have lots of memory and therefore need to page to disk).
-
- 3) How well does the operating system do things like cache
- recently accessed hard drive contents in RAM?
-
- Linux/X-windows loses to DOS big in (1). If svgalib supports svga
- modes on your card, Executor/Linux/svgalib should be about the same
- speed as Executor/DOS.
-
- Linux beats DOS in (2), at least if you're using cwsdpmi.exe (the free
- DPMI provider we supply). OS/2 Warp and other DPMI providers seem to
- do a substantially better job of using virtual memory that cwsdpmi, so
- if you're using one of those you won't see much difference moving to
- Linux. For you CS types out there, cwsdpmi uses a round-robin paging
- algorithm, an algorithm which doesn't work all that well for paging.
- It's on the author's long list of things to improve.
-
- Linux does a good job of (3), although something like smartdrv under
- DOS is probably similar.
-
- In general, Executor/DOS tends to be faster because Executor has
- direct video access (under many system configurations), and there's
- not much speed difference in the other areas. Your mileage may vary.
-
- - -Mat
-
- ------------------------------
-
- From: Jason Best <ocean@TSO.Cin.IX.net>
- Date: Sat, 4 Nov 1995 18:15:16 -0500
- Subject: totally unrelated question
-
- Cliff and co. - ignore this question...I know this is totally (well, mostly)
- unrealted to Executor, and prolly shouldn't be posted here, but oh well.
- Where exactly can I get Linux for an X86 machine, and how big is it?
- thanks anyhow...kinda curious to see how many OS's I can run on the same
- system I guess...:)
- later
- - -ocean
- (oh yeah - please reply via e-mail, as the list need not be cluttered any more
- with this string)
-
- ------------------------------
-
- End of executor-digest V1 #324
- ******************************
-
-
- To subscribe to executor-digest, send the command:
-
- subscribe executor-digest
-
- in the body of a message to "Majordomo@NACM.COM". If you want
- to subscribe something other than the account the mail is coming from,
- such as a local redistribution list, then append that address to the
- "subscribe" command; for example, to subscribe "local-executor-digest":
-
- subscribe executor-digest local-executor-digest@your.domain.net
-
- A non-digest (direct mail) version of this list is also available; to
- subscribe to that instead, replace all instances of "executor-digest"
- in the commands above with "executor".
-
-