True power consumption of laptops vs chartplotters?

[John_Clarke]

To add further to Tome's detailed figures here is some data for the C series.

I have a C120 and have measured the power consumption with the radar off (the radar comes on in standby automatically and it halves the power consumption to turn it off). When the C120 is on the following are also on – 2 No ST60 instruments, fluxgate compass and gyro stabilisation, NMEA/Seatalk interface and ST4000 autopilot on standby (all of these are powered from the Seatalk cable which is live when the C120 is turned on) and the Garmin 128 GPS (which has to be on to input to the C120). With all these on (they only add 0.19 A) and with full backlight the consumption is 1.45 A which at 12.5 v is 18.13 W (or 15.75 W for the C120 alone).

Raymarine claim 12 W at full brightness which appears to be a bit on the low side. Perhaps if run off 24 v the consumption is lower?

 

[npf1]

Tome - the sunlight readable PC is one of these http://www.xploretech.com/index.pl?iid=2233&isa=Category

 

[BlueSkyNick]

OK spill the beans, how much?

 

[npf1]

the tablet PC? About £600 - but I had to do some Dellboy style wheeling and dealing to on the way to achive that, otherwise would have been way out of my reach! Apart from the BSOD, which is a driver problem that I've not been able to resolve yet, it really is an excellent bit of kit. It gets used for much more than the nav software.

 

[Ships_Cat]

I think it was my previous post that has raised the question of notebook current draw again. I won't get into details again but a few points.

Those who give consumptions measured using a battery monitors such as typically found on pleasure vessels are likely to be well astray, one needs to measure the current draw directly, but that raises the matter of the draw fluctuates according to activity on the computer. Also. if one says that their notebook draws 60 watts when not charging, for example as some do, and then tells yer that their charger or other supply capable of charging is 60 watts, then they are obviously likely telling porkies.

The figures I gave in the other thread (and they are for several machines along the lines of your post) are based on direct current measurements, usually when running our ECS which has the hard drive writing all the time as it logs every NMEA sentence received and during under which the draw is quite constant.

A ready way to get a magnitude for the draw, as I set out before, is to take the capacity of the battery and divide that by the time in hours the notebook runs for before the battery is depleted. Typically, for the type of machine you mention they will have an around 4 AHr battery and run for around 2.75-3 hours when the battery is new ie it still has full capacity. Even if one assumes only a 2 hour run to battery flat to get a worst case indication of current draw, that means that the average amps drawn is in the vicinity of 2 Amps at the battery voltage for which is 10.8 volts if the typical 6 cell Li-Ion pack, so draw is a smidgin over 21 watts.

Important to realise that this magnitude approach assumes the battery was actually its rated capacity when new (would usually be expected to be less as Li-Ion cells age from the day they are manufactured) and one must use the run time for the notebook when the battery was new. But even playing around with the figures one can see that it always turns out to be much, much less than the 60-70 watts that many quote (usually off the charger).

Like all things there are heavy draw notebooks and also heavy draw plotters (the larger plotters, not intended for pleasure use, are basically PC's and have current draws to match). I am not talking about the exceptions.

The litmus test on our boat, which I quoted in the other thread and was measured as I wrote the post on the boat, was that with the notebook running as it was, the refrigeration running (but not the freezer), all instruments running (and they are consderable consumers on our boat - over 1 Amp), radio, minor consumers such as bilge alarms, Fixed Cellular Terminal, etc running, the total power drawn by the whole boat (determined exactly by measuring the current drawn from the boat's batteries) was about the same as what people popularly assume notebooks draw - without hauling out my previous post to double check, the whole boat draw at the time was around 70 W from memory.

Perhaps a point to come out of this is if people are looking at getting a notebook and power draw is important then they should stay away from high fangled specialist machines targetted at users where draw is not important. The simple check is to get the capacity of the battery for the intended machine and divide it by the expected run time (even using the manufacturer's usually inflated claim for this adjusted down to expected reality). Clearly, the same applies to plotters, high end ones can have considerable draws too.

I suspect, but have not checked, that mid spec machines aimed at the mobile business community are the least power hungry notebooks.

I use an ECS on a notebook for nav but it is not my intention to claim that is the ideal solution - that depends on the boat and the extent of the desired nav needs (for us official electronic charts are essentially free, and that is an added attraction to use of a PC). But the often quoted extreme power demands are very misleading to those trying to make a reasoned decision.

John

 

[AOWYN]

Which software are you using that loses contact with the dongle? I have been having the same problem with Seapro. do you know if there is a half-way house with only the power hungry processes, (the screen?) shutting down?

 

[Vara]

Can't you do that with "power options" in XP,or is that too basic an answer?

 

[npf1]

I'm using Seapro too. Don't know about a laptop, but on the tablet pressing the off button momentarily turns off the backlight.

 

[Sybarite]

Exactly my sentiments. I have two portable GPS's which I use to correct my dead reckoning. Makes navigating more interesting and means that when the power failure happens, panicking becomes optional.

John.

 

[Sybarite]

..."I've been a rugged tablet PC with a sunlight readable 10" screen with a pentium M processor for about 18 months..."

Were you a laptop before?

John

 

[tome]

John

I think that using battery capacity and run time is a bit of a red herring, for the following reason. Most laptops have quite sophisticated power management systems which will dim the backlighting and shut down the hard drive etc when running on batteries. They do this to achieve a reasonable battery capacity.

Although you can override some of these settings, others are hard-wired. I'm using an HP Compaq 6120 and as soon as I disconnect power, the backlight dims. When I use it onboard, I use a DC-DC converter so it sees power available and runs in the more power-hungry mode with full backlghting etc

Also, when you state that data is constanly being written to the hard disk, I believe you may find it's being cached and only physically written when there's a full sector, or on closing a file

 

[Ships_Cat]

Hi Tom

Ok - interestingly the notebook I was using the other day was a nx6120 (Centrino 1.8x Mhz). For current measurement I had the battery fully charged and in the notebook and all power saving features off and measured the current to the DC buck converter I was using to power it - I emphasise the "TO the buck converter" so at boat's battery voltage not the buck converter's 18.5v, boats batteries were floating at 13.2 volts as I posted in my post in the other thread.

As you say, when you remove external power the screen dims (but certainly only a little) but that represents a further power saving over the powered check I did.

Regarding the ECS, is definitely writing continuously to the hard drive - on all computers I have run it on one can see the drive light going and (on the noisier ones) hear the head. It is produced for small naval and commercial vessels and I assume that the NMEA and other data (warnings, alarms, manual notes) is written immediately for data security reasons.

The original claim in the other thread is that notebooks draw 60-70 W typically {EDIT: have checked, claim was actually "70+ watts"} and indeed in this thread have been figures such as 5 A quoted (so in excess of 60W). These are all too high for a charged notebook. Refer my litmus test where the whole boat with a number of other significant consumers running was only managing 60 - 70 W (think it was 70 but haven't checked the other thread {EDIT: have checked, was 80, included frig, but not freezer, and a number of other consumers as set out in my post}).

Again, the rough calc using notebook battery capacity and run time puts to bed the matter. Have a look at you 6120 - tell me how many amp hour it is (likely 4AHr if std pack) and what run time you averaged out of it when new, and calc the average current draw to deplete the battery in that time and multiply by the voltage (10.8 if the 6 cell pack) for watts. Is rough but I guarantee it comes out only a fraction of 60 W regardless of what power profile you have set.

For mine, I got near 3 hours run when new (still much the same) so assuming the battery was actually 4AHr then average current draw is 1.33A and at 10.8v that is around 15 W (actually will be a bit more as get around 2-3/4 hours, but obviously less if running movies or something). Now maybe the capacity is not 4AHr and maybe the run is less than 3 hours, but you have to manipulate the battery capacity upwards and/or the run time downwards into stupid territory to even get halfway to the 60-70 W claim.

The few others I have checked have been much the same. I still have an AMD 1.2 Gig Compaq here which I checked on the boat a couple of years back with much the same results - I could take that one back and do it again?

Appreciate your views.

John

 

[tome]

Hi John

I will try running the 6120 via the DC-DC converter later today to see what the actual current draw is at 12.0V

Meantime, I put the machine next to an E120 display and took a picture which shows just how bright the Raymarine screen is by comparison. In fact, it was difficult to get a shot which didn't over-expose the E120

Next, I took a shot of the E120 on it's own using manual shutter speed of 1/120 sec

I then displayed this picture on the screen of the 6120 and took a further picture. I had to use a shutter speed of 1/4 sec - which is why it's a little unsteady

This is why I wouldn't consider a laptop suitable for daylight use, although it's fine below.

I'll let you know the results of the power consumption test once I have a chance to get down the boat and borrow the Dc-Dc converter

 

[tome]

Ok, that's a tablet PC rather than a laptop but I cannot see any specs for the screen brighthness

 

[tome]

Ok, tried running the 6120 on DC power. The input voltage to the DC-DC converter is 12.0 Volts and the output is 19.0V (the standard power supply provides 18.5V). Incidentally, the battery capacity s 4.8 Ahr at 10.8V

It's difficult to measure the current on a digital ammeter as it jumps around a lot, but it appears that the minimum draw is 1.8A and this increases to 3.0A when doing any disk access, say a mean of around 2.4A or mean power consumption of 28.8W. This is certainly well below the 70W claimed by some. I believe this PC is fairly typical, so to put it into perspective:

1) A small dedicated plotter such as the Raymarine E80 draws slightly less power than a laptop PC even when running at full brightness. The C70 (which isn't as bright) draws sigificantly less power

2) The larger the screen area, the more power consumed. The E120 uses slightly more power than the PC at full brightness

3) A laptop is nowhere near as bright as a chartplotter and cannot be used in the cockpit on bright days

4) If the backlighting on the chartplotter is reduced (possible on all but sunny days), the power saving is significant and the net consumption considerably less than a laptop

5) If the chartplotter incorporates radar then the net power saving over separate screens will be significant

So I think we'll keep the laptop for passage planning and in-port entertainment, and use the chartplotter for charting/ AIS/ radar

Incidentally, we used the E80 on a recent passage to Ireland and back and both SWMBO and myself concluded it was just too small, so we're fitting the larger E120

 

[BlueSkyNick]

my two penny'th -

- wouldnt even contemplate taking the laptop into the cockpit, regardless of power consumption - simply too vulnerable.
- interesting to read how much a difference there is by varying the backlighting, very useful
- changing from E80 to 120 - you must be getting these things buckshee from somewhere! (or saving more than you though by not smoking!) /forums/images/graemlins/wink.gif

 

[tome]

Loaned by the good peeps at Raymarine, though I did have to bite the bullet when we opted for the E120!

 

[Ships_Cat]

Thanks for the effort in doing that Tom. Very interesting, also the comparison with the Raymarine display.

Absolutely agree about the sunlight on notebook screens - luckily (it was not planned like that when the boat was designed) I can see ours on the chart table from the helm position but not read the detail though, from companionway is fine. In my experience on commercial vessels visibility in wheelhouses is fine (although they all have had high performance window glass).

In my view peeps need to keep in mind what they are actually after, such as if a more powerful nav tool is wanted they need to move away from a simple 5-6 inch screen pleasure plotter but otherwise need not, how wet their boat is (usually a function of boat size), power capabilities, whether want to use official charts, sunlight viewable, cost, etc rather than the often simple (and wrong) dismissal given by some that notebooks are no good because they draw 60-70 watts or whatever so little 5 inch screen plotter is the only way to go.

For our own part a notebook or any similar power draw is of little consequence because even if left running continuously for 12 hours, which is most it takes us to get out of serious pilotage here, is prob only another 10-15 mins run on the engine charging, so suits us but others are different. I also like the ease and tools available on the ECS and as I think I said before our electronic official charts are essentially free and so need an appropriate ECS to run them, again affects ones decision (but we would stick to official anyway, again part of the decision). Our boat is also very dry (and walk past the bathroom to chart table so easy to dry hands and any drips on way to notebook). All things part of OUR OWN decision but not necessarily applies to others - I think some posters forget that needs, desires and situations are different but perhaps is only a skill gained though through hands on specifying of boats.

Now shall we get onto refrigeration and power /forums/images/graemlins/smile.gif?

John

 

[BrendanS]

The laptop and notebook manufacturers are very aware of screens and outdoors. HP have committed to have all their laptops to 500 to 5550 nits. My Asus 17" is 750 odd nits. A true out door sun on screen is about 1000 nits. Huge difference to the normal 200-250 nits of average laptop screen of 1 year ago

Most people have no idea, but are learning quickly

 

[Ships_Cat]

Me again /forums/images/graemlins/smile.gif.

May be of interest to some as gives a good idea of how power needs of a notebook vary according to activity and power management.

Most (all?) now manage the CPU clock to minimise power usage ie CPU clock is slowed if low processing needs and speeded up again if needed (they also manage the core voltage for same reasons).

Again, on a 6120 similar (not sure if same precessor tho') to Tome's I have done a check on that (tool was Lavalys). Core voltage claimed to vary 0.94 - 1.34 V and typical power (CPU only) 3 - 21 W - I believe the tool is quoting that from specs, not by direct measurement. So a very big range of CPU power demand depending on the tasks in hand.

Perhaps another aspect of minimising power then is the choice of nav program. The ECS we use is very oriented to function and while runs with Windows it creates its own user interface, etc (but can run Windows apps over the top of it). It is probable that it is less power hungry than some (many?) others using the Windows interface? (I am not recommending it as not aimed at pleasure boats, for example it does not read magnetic compass NMEA sentence, just gyro, just mention as an example.)

I have got it running on the same notebook here but with no NMEA input is hard to get an idea of processer load (I actually don't think that would make much difference, but logging to the hard drive is much reduced without it) but is quite low even if loading charts, dragging chart across the screen, doing nav things such as dividers, bearing lines, etc. I'll run it when I am back on board mid next week with NMEA inputs and full logging going.

So maybe ECS choice is another important aspect of power demand?

John