GlobalSat has released an update of their popular BU-353 USB GPS receiver. The new device looks and operates just like the old one.

The only difference in appearance is the old model has "GPS" on a white translucent strip on the top and the new one has G-STAR IV on a translucent black strip on the top.

The model number is now BU-353 S4.

The S4 in the model name is an indicator of the one significant change between the old and new models. The old model has a SirfStar III GPS chip in it. The new model has a SirfStar IV GPS chip in it.

While that's the only obvious difference between the two -- operationally they appear to be identical -- it's a really big difference in performance.


I just got a new notebook computer a few days ago so I set it up on the kitchen table to test the new receiver. I downloaded and installed the driver from GlobalSat's web site because I don't have a DVD drive on the new computer. I also downloaded and installed GlobalSat's little GPSInfo test utility which is always handy for configuring and monitoring a new receiver (works with any generic NMEA device, not just GlobalSat's).

Next I downloaded and installed Franson's GPSGate Express. I find it a handy tool to use whenever I have a new GPS receiver to test. I connected the S4 to the computer about the same time as I started downloading GPSGate. By the time GPSGate was installed and running, the brand new S4 already had a location fix.

That's impressive because it was just sitting on the kitchen table beside the notebook. A brand new GPS fresh out of the box needs to do a complete cold start to figure out where it is. This can usually take quite awhile and almost always needs to be done outdoors or it will take even longer or won't happen at all.

I already had Streets & Trips installed. It had no problem finding and using the S4.

I installed Street Atlas and it also found the S4 and started using it right away.

I haven't had a chance to try it in a vehicle yet but if it works that well inside the house I have no concerns about its performance in a vehicle.


...ken...

I've attached a screenshot comparing the performance of the S3 and S4 versions of the BU-353. I didn't have a program that would allow two instances to run with a different receiver connected to each. So the S3 is running in nRoute and the S4 is running in GPSInfo.

They both have similar satellite displays so you can see which satellites are being used and the signal strengths of them.

This is with the notebook sitting on the kitchen table and the two receivers connected to opposite sides of the computer.

As you can see, the S4 is doing much better than the S3.

It should be pointed out that even the S3 version is doing very well for being indoors.

...ken...

Attached Images

S3-vs-S4.jpg (56.9 KB)

Ken, these are similar preliminary observations to what I am seeing.

I note that you got the GPSinfo program working. I have not. Even though the GPS is registering in Streets and Trips, GPSinfo does not detect the BU-353S4 when I try to conduct a scan. Any idea what I am doing wrong?

In the above test, the red represents the old BU-353 S3 unit, and the blue is the new BU-353 S4.

The test was performed in downtown Toronto. The area was chosen because of its traditionally difficult environment for GPS navigation due to very tall buildings surrounding narrow streets.

Both receivers were placed on the roof of the vehicle, approximately 30 cm (1 ft) apart, and the testing of both units was performed simultaneously.

Notice that the blue line (the new S4) tracks the road more closely than the red line (the old S3). This difference can really be only observed in very difficult environments.

The new BU-353 S4 had a fix on one or two more satellites during the test than the S3 version.


The old version has received very positive feedback from many users over the past few years, some of which can be read here: http://www.laptopgpsworld.com/4800-globalsat-bu-353-usb-gps-receiver

The new version of the receiver works just like the old version, and it even adds a bit more precision when used while navigating in difficult environments. This may be partly due to the improved sensitivity of the SS4 chip, -163dBm, versus the SS3, -159dBm.

Specifications:
Several users of the new S4 version of this unit have commented on the faster start up time. This may be due to the following new implementation:

Anyone know if the S4 chipset is intrinsically more expensive than the older S3 was when it was new?

Or is the nearly double price for the BU-353S4 vs. the older BU-353S3 just due to the older model now being a deeply discounted end of life product?

BU-353 S3 $28.95
BU-353 S4 $56.95

First SiRFstar III based GPS receivers started to ship in 2005 (the GlobalSat 353 was one of the very first ones), so my guess is that the SiRF4 chip itself is unlikely more expensive now than the SiRF3 was years ago. If anything, it is likely less expensive (again, only a guess). So we are just seeing discounted price of the old stock, and a higher price of a brand new product. Those who cannot wait for the latest and greatest will pay a price premium. Those who are patient to wait a few months for the price to drop will save a bit.

[EDIT] I just found some historical info from December of 2007. At that time the old S3 version sold for Regular price: $79.99 Sale price: $45.99

Mine just worked. I didn't even have to tell it to scan. It already had the correct settings selected so I just had to poke the "Start GPS" button and the sentences started to scroll.

Oh..... memory is starting to trickle back.... I first tried to download GPSInfo from their website when I downloaded the drivers (no DVD drive on the notebook). But the file seemed to be broken. The drivers worked fine but the GPSInfo download was really short and when I tried to open it it tossed an error that it was an invalid file. I tried downloading it three times with the same results.

So I broke down and dug out my portable DVD drive and installed GPSInfo from the mini CD that came with the device. Installation was clean and it ran first try as described above.

...ken...

Thanks, Ken. I have tried on my stationary PC both the GPSinfo.exe file that came on the CD and the downloaded one with the same result: the app runs but scans do not reveal any port with a GPS on it. I tried it on my laptop, PC, and UMPC all with the same (non-)result.

However, this is a trivial inconvenience inasmuch as BU-353-S4 works just fine with S&T on all three devices.

Hi. Am I seeing those images correctly? Does the S4 put you in the middle of the northern Pacific?

That is a representation of the satellite locations! Last time I looked, Regina SK wasn't in the Pacific.

Terry

Good one, mate!! Terry is right.

It doesn't get much drier in Canada than here in the middle of the Canadian Prairies.

..ken...

... but, if the second ring represents the horizon and the centre is your current position?? I'll make sure to check the coords next time. Cheers.

My S4 arrived today so I put it to the test against a BU-353 (S3) on a computer running Win 7 64. A couple of things I noted;

1)You don't need the CD as the appropriate drivers will be downloaded from Windows.
2)The copy of GPS Info on my CD is old and won't work with Win 7 64. It can't find any COM ports at all.

I downloaded the version at GlobalSat support and it worked great.

http://www.usglobalsat.com/s-24-support-drivers.aspx#A

The S4 was considerably better than the S3 in the location I tested. I also tried the Cold Start button in GPS Info with the S4 and had lock in 45 seconds at this location. For easy comparison, I just opened 2 instances of GPS Info and pointed each at the appropriate COM port (7 and 9 in my setup).

I wiill test it against a MTK I bluetooth GPS later and expect it will at least equal the performance of that chip.

Terry

Attached Images

353 Comparison.jpg (91.2 KB)

Thanks for the update and the link, Terry.

The copy of GPSInfo on the CD with my S4 works fine on Win7 Home Premium on my new notebook. But it won't let me launch a second instance. I'll download the one from your link and see if it works better for me.

Re Windows loading the driver when you plug in the S4, my notebook threw an error when it tried to do that. That's why I downloaded it from GlobalSat.

What's the cold start time for the S3 in that same location?

...ken...

While I was away, the BU-353 finally got a fix! I have tried a cold start at this location but can't outwait it...something like 4 minutes. Strange, as I normally depend upon it here.

It wasn't clear whether the MTK1 or BU-353 S4 was a winner. They both gave similar results but the bars were slightly longer for the MTK. Unfortunately, I am unable to use the Cold Start button successfully with the Bluetooth GPS.

Terry

Attached Images

353 Comparison_2.jpg (101.2 KB)

S4-MTK1 Comparison.jpg (106.8 KB)

I realized I also have a MTKII chip GPS, the iBlue GM-2, another USB device. I did a quick comparison and would have to say the BU-353 S4 is superior. This may be result of the GM-2 design though, rather than the chip itself. Still, it shows the BU-353 S4 to be a great unit.

Again, I was not able to use the Cold Start feature!

Terry

Attached Images

S4-GM2 Comparison.jpg (104.8 KB)

Just in case it is not clear where I am, here is the S4 and S&T (2004?)!

Terry

Attached Images

s4 S&T.jpg (81.0 KB)

Funny! Even with the GM-2, although Windows didn't have a driver, it directed me to a site to download what was needed.

Maybe I'll set it up again and try to get some screen captures of the process. As its a Prolific driver, I only need to move it to a new USB port (if I have any left).

Terry

My first cold start was about 45 seconds and everything else since then has been 'warm' and about 2~5 seconds.

My problems reported earlier with the GPSinfo.exe program seems to be that the version provided on the in-package mini-CD is probably 32-bit only. Installing the version available from the website that Terry provided the link for above, gives me a working GPSinfo.exe file on my 64-bit PC.

I ran the driver test this morning but used a different computer to ensure the driver wasn't already present. I connected the BU-353 S4 to my Fujitsu Win 7 32 slate as it has never had the Prolific driver installed. Still, the identification and driver loading went very smoothly.

I used PicPick to ensure I could quickly get a series of screen captures. Other than that, I only had to click on 'Click here for status' to allow the other screens.

Terry

Attached Images

353_1.jpg (59.1 KB)

353_2.jpg (91.0 KB)

353_3.jpg (90.9 KB)

353_4.jpg (65.0 KB)

353_5.jpg (75.1 KB)

I figured out the driver issue. It was a setting on my new computer that was stopping Windows from using Windows Update to search for a driver if there isn't one already on the computer.

Problem solved.

... Sort of. When I plugged in the GM-2, Windows didn't find a driver even when it was allowed to hunt the internet for one so I had to resort to a manual install of the CP210x driver. (I used the one for the i.Trek M7.)

I downloaded GPSInfo from the link you provided above. It's a different version than what's on the miniCD that came with the S4. I like that the information display is now all on one screen.

I was able to connect two GPS receivers to the notebook and run two instances of GPSInfo side by side for direct comparison which I could not do with the other version. Another problem solved.

I pulled out the Transystem GM-2 (MTK chip) for some comparisons with the S4 (SirfStar IV chip). From a strictly performance perspective, there does not seem to be anything to choose between them at my kitchen table. I'll test them in my basement office this afternoon. I'm pretty sure the GM-2 can get some kind of location fix down there. We'll see if the S4 can, too.

The main difference between the GM-2 and the S4 is the usability. And it's a huge difference

GlobalSat made a very wise choice leaving the serial communication set at 4800bps. That's a setting that provides sufficiently quick position updates for navigating anything except maybe a jet fighter and it works immediately with any navigation software.

The GM-2 defaults to 115200bps. That's going to be a problem for some folks. You need to know that the reason things aren't working is probably because of the COM port speed setting. And you need to know what to do about it. And you need to have a way of dealing with it.

Even some programs that will scan for active COM ports will not scan the port speeds up to 115200. GPSGate is one of the smartest for finding your GPS receiver and even it won't find the GM-2. You have to manually set the input COM port speed to 115200 before GPSGate will use it. Then you can use GPSGate to throttle it down to 4800 for programs like Streets & Trips.

So with minimal performance difference between the two receivers, the S4 wins hands down for immediate usability.

In my view, anyway.

...ken...

I agree wholeheartedly Ken. And I recall the original software that came with the GM-2 wouldn't change the bit rate even though it is possible with an older software version.

For my screenshot, I didn't even bother to track down the appropriate software ...just ran GPS Info at 115,200 for that instance.

Terry

The GM-2 can provide 5 fixes per second. That's probably why the baud rate is 115200.

--- CHAS

I think Ken's point was the GM-2 should ship preset to 4800bps to satisfy the largest number of customers out of the box. IIRC they did with later shipments.

What I didn't like was the software shipped with it did not allow one to change the bit rate while a previous version we had from our iTreks did!

Terry

I have had a chance to do some field-testing of my new BU-353S4 now and I thought I would share some of my observations. First, I should make clear that I cannot compare it to an S3 version because I do not have one. So, my personal benchmark would be the GPS receiver that is built into my Viliv S5. Second, I should point out that this is in no way a balanced test; all terraines, all conditions, etc. However, I did use it in a fairly hostile environment: Manhattan. I say "hostile" because the tall buildings we have here are not friends of GPS. When they are not busy blocking out the GPS signal, they are distorting it as the signals bounce off their large, flat, vertical surfaces.

The S4 definitely out-performs the Viliv, however. An easy indicator is how many satellites are being recognized. For instance, parked between two 50-story towers, the Viliv showed 4 where as the 353S4 showed 7, 8, or 9.

The S4 produces a relatively true and straight GPS trail that seems to be pretty accurately positioned. You can even see where I had changed lanes. No, I did NOT cut off that taxi! I was there first. Even on relatively wide streets with low-ish buildings, I would get a sort of saw-tooth effect on the GPS trail with Viliv as the recorded trail was distorted by varying signal qualities.

The start-up seems almost instantaneous. After being on standby and resuming the GPS navigating, I had a red dot before I could get the key in the ignition. This is a major boon.

Also, the location lag seems to be substantially reduced. In other words, the red-dot or arrow in Streets and Trips is much closer to where you actually are rather than where you were 5 or 10 seconds ago. This can be a real help in giving driving directions in unfamiliar territory although I have not actually tested that.

So, in summary I am favorably impressed. I cannot think of any negatives. Get one!

SF, I have the S3, with which I am thoroughly pleased. I understand that you do not have my model, so you can not make a comparison between the two as to tracking lag. But, I would appreciate it if someone else could comment on what I should expect pertaining to lag... About the only incentive I would have to upgrade would be a reduction in lag... Thanks...

My understanding is the Viliv internal GPS is based on SirfStar III technology so it is a fairly reasonable comparison. However, as I cannot think what would cause the lag, I can't see how a more sensitive GPS would reduce it.

Terry

Given the potential limitations of antenna size and positioning possibilities between an external device like the BU-353 S3 and the internal S3 chip in the Viliv, I would not assume there is any basis whatsoever for comparison. In fact I would fully expect the internal chip in the Viliv to perform much worse in almost any test, versus the BU-353 S3, because of its severe limitations.

That is not a criticism of the internal chip in the Viliv. It is simply an observation of the limitations of any built-in GPS chip, in PCs, tablets and phones.

Regarding the lag issue, a GPS chipset will not deliver the necessary position sentences to update a position until it has determined a position it is confident to be stable. In a limited or impaired reception situation the chip may be doing multiple position fix acquisition and analysis calculations for any single position delivered to the receiving/display application. This could result in varying amounts of apparent "lag" in the position display.

Or the error factors being delivered with a position fix (HDOP, etc.) may be sufficiently bad that the receiving/display application may simply reject many of them and may only update the position display when it receives fix updates within an acceptable range of error.

...ken...

I agree with you on placement issues Ken. The S3 would have a better sky view than the internal GPS of the Viliv depending on placement. However, I find your lag explanation to be based on a few too many assumptions we are not able to prove to state the S4 would reduce it.

Maybe I'll have to try S&T with each to see. But I am somewhat limited to only good reception areas while travelling!

Edit: Marvin, you have both and access to a large metropolitan area. How about a hands on!!

Terry

I'm not sure I understand, Terry. Are you questioning SpadesFlush's statement that he observed less lag between the Viliv internal GPS and the S4?

Or just not sure whether my guesses about the possible causes of the difference he's seeing hold water?

...ken...

I find it hard to believe that a better GPS reduces lag. I have always suspected lag (which I have never experienced as I so rarely use S&T) is a function of the computer or the software being too busy doing other things to bother updating the position. Any GPS chipset operates at a bit rate allowing it to calculate position more than once a second...it is only that it provides the update once a second. So it is likely to produce something (as evidenced by SF's sawtooth track and one that Marvin produced years ago).

I also don't believe that S&T is smart enough to reject faulty locations based on HDOP (again as evidenced by SF's sawtooth).

Of course, these are only the opinions of a person who never uses S&T except for trip planning!!

Terry

Terry, I'm planning to venture to downtown Toronto for a side-by-site S4 vs S3 comparison this Saturday. Then I will edit my earlier post above by inserting there the test results.

@ Ken. Until I used the 353S4, I was reasonably happy with the GPS reception on the Viliv. I do not think it has anything to do per se with whether the antenna chip is built in or not except that one might have a bit more positioning flexibility with a 1.5m cable. For instance, I placed the S4 way up front where the glass meets the dash, whereas I usually would have the Viliv about 2 inches from the steering wheel.

@ Terry. I too thought the Viliv has the Sirfstar III chip, but I did not want to make that assertion without researching it. So long as it is reasonably well isolated from the other internal electrics (as it seems to be) I cannot see why it would function better on a tether.

@ Terry. As I tried to convey in my post, it was my impression only that there was a shorter positioning lag. I think there could be a logical explanation for that along Ken's lines. I am happy to stimulate the controversy so that perhaps the Forum can develop a concensus view on the subject based on a broader experience base.

Wow, you guys are just great! What great information! What excellent research... chipset verification, road work, live comparisons, etc... Thanks!

I just posted test results of NEW VERSION vs OLD VERSION. I edited my post above (post #4) by adding a screen shot and some comments.

Thanks for the test Marvin that more or less conforms to my experience. I would expect for most non-urban users (probably most users) that these improvements alone would not be a sufficient basis for up-grading from the S3 to the S4. However, other considerations might do the trick, such as shortened start-up times or, if in fact it is there, reduced navigating lags.

I tried to see if I could experience lag running S&T 2011 on my Fujitsu Slate. It has about the CPU power of a netbook and I loaded it with GPS Gate, GPS View, 2 instances of S&T 2011 and a screen capture program. I had to use an unpowered USB hub to connect the S3 and S4 as the slate only has one USB port. I assigned the S3 to GPS Gate and used its ports for GPS View and one instance of S&T; the S4 was connected to the other instance. The computer was certainly taxed but seemed equal to the task.

I took a 40 minute trip with the screen capture program set to 20 seconds. Unfortunately, I was unable to determine anything about lag or the differences of the GPS. From the screenshots, occasionally one was leading slightly and at other times the other but the differences were miniscule. Sometimes the S4 showed more satellites and sometimes the S3.

The difficulty is there are no problem areas for GPS around here except the park this side of Victoria. I didn't want to drive that far just for a test so the worst I saw was 5 satellites and usually 8 or more.

As I pushed the computer hard enough that Win 7 eventually complained, I can no longer suspect lag to be due to this. However, until I can see it, I'll stay out of the discussion!

Terry