Re: Garmin 2610 -- SHOCKING results from "slow" 1GB CF card
- From: racerx000@xxxxxxxxxxx
- Date: 23 Jun 2005 19:43:32 -0700
I've found that Lexar CF cards outperform Sandisk in my Nikon digital
camera as well. I have a regular Lexar 128MB CF and a 12x Sandisk CF
and Lexar writes faster in my Nikon Coolpix 4300 with flash on. I think
Lexar is just better.
>> Garmin 2610 -- SHOCKING results from "slow" 1GB CF card
All 5 messages in topic - view as tree
Bob Pariseau Sep 25 2003, 1:31 pm show options
Newsgroups: sci.geo.satellite-nav
From: "Bob Pariseau" <bparis...@xxxxxxxxxxx> - Find messages by this
author
Date: Thu, 25 Sep 2003 13:30:19 -0400
Local: Thurs,Sep 25 2003 1:30 pm
Subject: Garmin 2610 -- SHOCKING results from "slow" 1GB CF card
Reply to Author | Forward | Print | Individual Message | Show original
| Report Abuse
I just completed tests with a "slow" 1GB CompactFlash card to compare
with
the "fast" card I reported on yesterday.
The "fast" card is the just released SanDisk Ultra II 1GB CompactFlash,
rated by SanDisk as having a Read speed of 10 MBytes/Sec and a Write
speed
of 9 MBytes/Sec -- i.e., a "60x" card. Cost delivered = $260.
The "slow" card is a Lexar 4x 1GB USB-enabled CompactFlash purchased
about 2
months ago at CompUSA for around $220. The "4x" rating, according to
Lexar,
means the card has a maximum sustained Write speed of only
600Kbytes/sec.
The Read speed is not specified by Lexar. This card is NOT labeled as
having Lexar's new "write acceleration" technology. The "USB-enabled"
labeling means that Lexar has included USB code in the on-card
controller,
so that the card could, technically, be connected directly to a USB 1.1
port
for reading and writing using only a cable. Of course a standard USB
cable
won't work since the card end has to attach to the card's edge
connector.
But Lexar sells a "JumpShot" cable for just this purpose. The JumpShot
cable works ONLY with "USB-enabled" Lexar cards, and frankly the point
of
all this escapes me since the JumpShot cable is not significantly
cheaper
than a name-brand USB CF card reader/writer which will work with ALL
brands
of CF cards.
If you read my post yesterday, you'll remember that I reported that I
could
write map data to the "fast" card using a PCMCIA (pc-card) slot adapter
at a
rate of a little over 1.3 MBytes/sec -- substantially slower than the 9
MBytes/sec at which the card is spec'ed. The assumption was that the
PCMCIA
slot in my laptop was the limiting factor here. I also reported that
the
"fast" card produced no significant performance gains when used in the
2610
compared to SanDisk's previous generation 256 MB "Ultra" card (rated at
under 20x speed) which I had been using previously, and that there also
seemed to be no performance LOST in the 2610 due to bumping the on
board
data up from 256MB in the Ultra card to 1GB in the Ultra II card.
What I was expecting to find with the Lexar 4x card was that it would
take
about twice as long to transfer map data to the card (the 600KB/sec of
the
card would now be the limiting factor), and that performance using the
card
in the 2610 would be unchanged or perhaps a bit slower.
Boy was I wrong!
First the transfer time to load map data onto the card.
The "slow" card presented itself to Windows as having 982 MB "free
space" as
compared to the 977 reported by the "fast card". As has been
previously
reported, there is variation in the actually usable space on these
cards
that are all sold as 1GB cards. Since the new "slow" card was bigger,
I was
able to use precisely the same map set for testing as I reported on
yesterday.
That map set contains 676 map chunks from City Navigator V5 covering
basically the eastern half of the US and Canada, plus the Hawaiian
Islands.
MapSource adds the map data up as 976.6 MB. However the actual data
transferred to the CF card after the files are built only adds up to
946MB.
I used the same laptop as yesterday.
After the computer has chugged for nearly 1 hour 45 minutes building
the
index and map files, the actual transfer of data to the "fast" card
took 12
minutes -- a rate of 1.3 MB/sec.
The transfer to the "slow" card took 13 minutes -- a rate of 1.2MB/sec!
Clearly the Write speed of the "slow"card is greater than the 4x spec
Lexar
puts on it.
I can imagine two possible explanations:
1) Lexar is no longer making cards using it's older, slower technology
and
either cherry picks it's production runs so that the best cards are
labeled
40x and the ones with a problem here and there meeting all the specs
are
down-labeled as only 4x -- or perhaps they just label some that way
solely
for marketing reasons to send supposedly "different" products to the
discount retailers, or
2) I got lucky and ended up with a mislabeled card.
Whatever the case, it was a pleasant surprise.
I then took the "fast" and "slow" cards, each containing the same data,
out
for some tests in the 2610.
The 2610 used for these tests has been upgraded to firmware V2.40. All
route creation tests were done outside with the GPS turned on and the
vehicle stationary. The route preferences were set to calculate the
"Best"
route for a car with all the avoids (U-turn and such) turned OFF.
Each of the timings reported below are an average of 2 or 3 tries with
each
CF card.
1) An 11 hour route from NC to Cafe Du Monde, New Orleans. "Fast" =
44sec.
"Slow"= 29sec!
2) A 12 hour route from NC to a waypoint in Greenwich Village,
Manhattan, NY
(695 miles, 11:18 driving time, 18 turns generated for the routing
list).
"Fast"= 1min 44sec. "Slow"=38sec!
3) A 12 hour route from NC to farm land (only local surface streets for
miles in any direction) along the NY, PA border. "Fast"=46sec.
"Slow"=29sec!
4) A 12 hour route from NC to the suburbs north of Philadelphia.
"Fast"=1min 25sec. "Slow"=36sec!
What a difference!
Next I looked at the time involved in looking up things in the database
of
points of interest.
The first thing I noticed was that the lookahead speed while typing in
names
was NOTICEABLY faster when using the "slow" card.
But here's a more objective measure.
Given this map set, press Find/Lodging/Hotel Motel/By Name/Spell. Type
in
"Ritz". At this point the lookahead shows "Ritz Carlton Amelia [and a
bit
more I don't remember]" and says there are 25 matches in the list.
Press
OK. In the list presented, a few entries down we find "Ritz Carlton
Hotel"
indicating "4 entries". Press that name.
The 2610 will now search it's database to pull up just those 4 entries.
For
whatever reason, the algorithm used at this point takes some time.
On the "fast" card" it takes 30sec. On the "slow" card it takes 11
sec!
Clearly the "slow" Lexar card provides significantly greater
performance
when doing these data heavy tasks on the 2610.
I did not notice any significant change in operations during driving
(screen
redraw and such) between these two cards. The operations were "fast
enough"
with both of them.
The upshot here is that there CAN be a SIGNIFICANT performance
difference in
the operation of the 2610 depending upon the choice of CF card. I
don't
know whether my "slow" CF card was simply mislabeled, or whether the
USB
controller inside the card somehow made it easier for the 2610 to
access
what it needed to access, or what.
I also have no idea whether a Lexar "40x" card, the fastest they are
currently shipping, would have any significant performance difference
when
compared to this "4x" card.
So far all I know is that one specific "slow" Lexar card beats the
pants off
two, supposedly faster SanDisk cards when used in my 2610.
--Bob
.
- Follow-Ups:
- Prev by Date: Re: TomTom 5 Upgrade - First impressions
- Next by Date: Re: Garmin 2610 -- SHOCKING results from "slow" 1GB CF card
- Previous by thread: Garmin GPSMAP 296 Terrain Aviation Database Update for FREE
- Next by thread: Re: Garmin 2610 -- SHOCKING results from "slow" 1GB CF card
- Index(es):
Relevant Pages
|
