Article View: comp.sys.ibm.pc.hardware.chips
Article #595Re: Is it possible to actually get a 16 MB/sec data transfer rate w/ an EIDE drive?
From: Mark Booth
Date: Mon, 22 Dec 1997 00:00
Date: Mon, 22 Dec 1997 00:00
104 lines
5907 bytes
5907 bytes
In article <01bd0caf$43fe6ec0$0500a8c0@k6-233>, Scott Cooper
<seehorse@earthlink.net> writes
>Yes, but you're not talking about professional quality results, you can't
>get 6MB/s or higher video out of an EIDE drive, that requires SCSI
>(although it doesn't need UW, Fast SCSI-II is adequate).
I think that you are a little out of touch. The benchmark below is from
a new Quantum Fireball ST3.2A that I have just fitted to my machine here
at work. As you can see only the last 10% of the disk falls below 6MB/s
sustained transfer rate.
By the way, this a one hundred and eight pound (108UKP) drive. So this
drive is now considered 'entry level' by system integrators.
The test was done using the integrated EIDE of Asus TX97 motherboard.
>For hobbyists and
>home video, 3MB/s or less is adequate, and the better EIDE drives are fine.
> A professional Non-Linear Editor is a very time-critical application, and
>I know of none that don't use SCSI.
However remember that the FB ST 3.2 is not an AV drive, so I expect it
will do the occasional thermal recalibration - not ideal for direct
video. All I wanted to point out is that IDE drives with similar
capacity and spin speeds will have similar sustained throughput.
The big difference between SCSI and EIDE is that EIDE drives only come
in one flavour - the cheap kind. At the moment that means 5400rpm max.
If you want a 7200 or 10000 rpm drive, you have to get a scsi drive -
its the only option.
In general I find that SCSI is a 20%-30% premium over identically speced
EIDE drives, thats your entry fee. Then you pay more for the higher spec
(spin speed/seek speed/AV spec/large cache) that avalable only to SCSI
interface devices. For most people SCSI is not necessary - but if you
must have (b)leading edge technology, and you are prepared to pay the
price premium for it - SCSI is undoubtedly technically better, just not
better in terms of price/performance.
Take care peoples, and have fun,
Mark.........
PS I've just had a play with a Maxtor 84320D4, 4.3Gb EIDE drive, which
benchmarks between 11.80 MB/sec sustained, down to 7.40 MB/sec, with an
average of 10.40 MB/sec, but on a new Dell M233s (bottom of the Dell
range, VX chipset motherboard).
-----------------------------------------------------------------------
Check Harddisk v0.99 (C)1996 Marnix Timmermans All rights reserved
Tester: BOOTH ( MB )
Tested on: 19-Dec-1997 18:31 with version 0.99
Computer: Intel Pentium running DOS (7.0)
------------------------- LOGICAL INFORMATION -------------------------
Name hard drive: QUANTUM FIREBALL ST3.2A
Serial number: 853727596923
Firmware revision: A0F.0800
Buffer size: 163 sectors (81.5 KB)
Max. and current blockmode: 16 / 16 sectors per interrupt
---------------------- BUFFER TO HOST THROUGHPUT ----------------------
1 2 4 8 16 32 64 127 (Blocksize in sectors)
3.6 5.9 8.9 11.5 13.5 14.5 15.0 15.2 (Throughput in MB/sec)
------------------------- PHYSICAL INFORMATION ------------------------
Total size: 3224.57 MB
Number of heads / platters: Unknown. Probably a 2-platter disk
Defect mapping: Unknown
Random seek: 9.5 ms
Random seek (first 512MB) 5.5 ms
Head / Cylinder switch: around 3.8 ms
Spinrate / latency: 5400 RPM / 5.6 ms
Performance index: 3.65 (max) 2.80 (avg)
Internal Sustained
part relative speed SPT Throughput Throughput
0% XXXXXXXXXXXXXXXXXXXXXXXXX 100% 273 12.60 MB/sec 9.35 MB/sec
10% XXXXXXXXXXXXXXXXXXXXXXXXX 100% 273 12.60 MB/sec 9.35 MB/sec
20% XXXXXXXXXXXXXXXXXXXXXXXXX 98% 268 12.35 MB/sec 9.20 MB/sec
30% XXXXXXXXXXXXXXXXXXXXXXXX 97% 264 12.15 MB/sec 9.05 MB/sec
40% XXXXXXXXXXXXXXXXXXXXXXXX 94% 257 11.85 MB/sec 8.80 MB/sec
50% XXXXXXXXXXXXXXXXXXXXXX 86% 236 10.85 MB/sec 8.10 MB/sec
60% XXXXXXXXXXXXXXXXXXXXX 83% 226 10.40 MB/sec 7.75 MB/sec
70% XXXXXXXXXXXXXXXXXXX 77% 210 9.70 MB/sec 7.20 MB/sec
80% XXXXXXXXXXXXXXXXXX 73% 199 9.15 MB/sec 6.85 MB/sec
90% XXXXXXXXXXXXXXXX 66% 180 8.30 MB/sec 6.15 MB/sec
100% XXXXXXXXXXXXXXX 59% 160 7.35 MB/sec 5.50 MB/sec
avg. XXXXXXXXXXXXXXXXXXXXX 85% 231 10.65 MB/sec 7.90 MB/sec
Max. Buffer to host throughput 162% 15.20 MB/sec
The interface is FAST ENOUGH for this drive.
--
Mark Booth
Technical Details
Message-ID:
<VCVhRNAHCun00AiH@oxim.demon.co.uk>Path:
arc1-ca.pugleaf.net!unknown!not-for-mailReferences:
<348c422b.31666353@news.demon.co.uk> <881691095snz@inf-eden.demon.co.ux> <pLMYl5dhX7hK-pn2-bmRtSYdnH0CM@ALEX> <3498a647.4325910@news.ebtech.net> <pLMYl5dhX7hK-pn2-SPgxduDgGt8b@ALEX> <67d4es$ntk@mtinsc03.worldnet.att.net> <01bd0caf$43fe6ec0$0500a8c0@k6-233>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 1036bc,aad589958d8c66fd X-Google-Attributes: gid1036bc,public X-Google-Thread: f4d7c,aad589958d8c66fd X-Google-Attributes: gidf4d7c,public X-Google-Thread: f5f89,aad589958d8c66fd X-Google-Attributes: gidf5f89,public X-Google-Thread: 109e38,aad589958d8c66fd X-Google-Attributes: gid109e38,public X-Google-Thread: 1045b9,aad589958d8c66fd X-Google-Attributes: gid1045b9,public From: Mark Booth <markb@oxim.demon.co.uk> Subject: Re: Is it possible to actually get a 16 MB/sec data transfer rate w/ an EIDE drive? Date: 1997/12/22 Message-ID: <VCVhRNAHCun00AiH@oxim.demon.co.uk> X-Deja-AN: 309440389 Distribution: world X-NNTP-Posting-Host: oxim.demon.co.uk [158.152.26.40] References: <348c422b.31666353@news.demon.co.uk> <881691095snz@inf-eden.demon.co.ux> <pLMYl5dhX7hK-pn2-bmRtSYdnH0CM@ALEX> <3498a647.4325910@news.ebtech.net> <pLMYl5dhX7hK-pn2-SPgxduDgGt8b@ALEX> <67d4es$ntk@mtinsc03.worldnet.att.net> <01bd0caf$43fe6ec0$0500a8c0@k6-233> Organization: Oxford Intelligent Machines Ltd MIME-Version: 1.0 Newsgroups: comp.sys.ibm.pc.hardware.chips,demon.tech.pc,alt.comp.hardware,comp.sys.ibm.pc.hardware.storage,comp.sys.ibm.pc.hardware.misc