[Info-vax] VSI has released 9.2-1

[kemain.nospam at gmail.com]

> -----Original Message-----
> From: Info-vax <info-vax-bounces at rbnsn.com> On Behalf Of Arne Vajhøj via
> Info-vax
> Sent: Friday, June 16, 2023 8:17 PM
> To: info-vax at rbnsn.com
> Cc: Arne Vajhøj <arne at vajhoej.dk>
> Subject: Re: [Info-vax] VSI has released 9.2-1
> 
> On 6/16/2023 7:05 PM, David Jones wrote:
> > I re-built SQLite and ran its speedtest program. The CPU time was
> > pretty much exactly half the previous run on 28-MAY-2023. The
> > benchmark is I/O bound, so the I/O processing overhead appears to be
> much improved.
> 
> Nice.
> 
> > Running a re-built Bytemark program under the new release is a wash,
> speedwise.
> 
> That is CPU bound right?
> 
> So it is sort of expected that it depends more on compiler than on OS.
> 
> Arne
> 

Might not apply here, but in other vendors FT versions of code, it was (is?)
common to leave debug code in for better support in resolving functionality
issues until the release became official. This would typically make FT
versions slower than the final release.

Walk down memory lane - Microsoft left the debug code in PROD versions of
Alpha Windows NT Office and the Alpha Windows benchmarks were still
significantly better than the x86 versions of the time.

We (Digital NT Wizards program at DEC) never could convince MS to release
Prod Alpha NT versions without the debug code.

Ah well .. and the rest is history.


Regards,

Kerry Main
Kerry dot main at starkgaming dot com






-- 
This email has been checked for viruses by AVG antivirus software.
www.avg.com