CPU heat & SETI
Peter Smithson
Hi,
From the SETI FAQ we've got -
"
Will running SETI@home overload or burn out my CPU?
No. The CPU on most computers is always executing instructions (often the
operating system's "idle loop") whenever the computer is turned on. It's
no additional strain to execute SETI@home.
"
What are the computers that don't fit into "most" ? (How can they say
"No" and then contradict themselves in the next sentance? Should be
"Probably not").
A friend showed me that when he runs SETI on his laptop, the fan will cut
in within 10 seconds or so. I understand some laptops have power saving
features so I guess that's the case here. He's got an Intel Pentium of
some sort in a Dell laptop. My Dell laptop with a slower processor
doesn't do the same thing - so I'm not sure.
Reading about one of the add-ons at http://www.suggestie.nl/seti/ it says
"The Seti client runs at low priority in order not to interfere with
other software. It however uses all CPU power you have left at any time,
which results in a constant 100% CPU load of your computer. This may
result in overheating problems which may damage your pc."
So what's the story?
I'm interested in the effects on -
Intel Dell Laptops
AMD/Intel PC's running Windows or some UNIX variant.
Itanium 2 with HPUX
Power PC - AIX.
Cheers
Peter
Peter Smithson
geo...@yourplace.com says...
> The fan cuts in because the alptop is using its cpu and therefore
> creating heat that then needs to be expelled, that is what the fan
> does.
Thanks - I know what a fan is for. Why is the laptop generating the exta
heat is my question.
The FAQ say this
"The CPU on most computers is always executing instructions (often the
operating system's "idle loop") whenever the computer is turned on. It's
no additional strain to execute SETI@home."
I realise it won't do the computers any harm to process data - that's
what they were built for. It's just that the FAQ entry suggests they
would use the same amount of power / generate as much heat if they were
not running SETI. Yet other evidence (including yours) suggests
otherwise.
Thanks.
Peter
Peter Smithson
I've done some reasearch on the web and I'm now pretty sure that the
increased CPU heat only applies to laptops with AMD or Intel chips.
Something to do with a HLT instruction which has been around for some
time. I got this info from the following links -
http://grassomusic.de/english/amdk7.htm
(Details more than anyone could possibly ever want to know about it!)
http://www.tomshardware.com/mobile/20030102/compaq_evo_n1015v-01.html
(a smaller summary)
http://www.microsoft.com/downloads/details.aspx?FamilyID=2898f8dd-10f8-
4107-9f7b-16c5a525de1e&displaylang=en
That last one just explains that the software side of things is catered
for in Windows XP. I also saw fixes for Linux to cater for the HLT
instruction.
Maybe laptops with other processors do something similar but I don't have
access to any of those so I'm not too interested.
So I think the SETI FAQ reply is accurate since most computers are not
laptops. Although thinking about it - the reply doesn't really relate to
the question - the question is something along the lines of "Will running
SETI blow up my computer". The reply is "No" but then it goes on to talk
about "additional strain" which is another matter.
The only other possibility is that some chip & OS manufactuers have
thrown this functionality into their non laptop computers without making
it clear - though I really doubt that.
Peter
Martin 53N 1W
> I've done some reasearch on the web and I'm now pretty sure that the
> increased CPU heat only applies to laptops with AMD or Intel chips.
> Something to do with a HLT instruction which has been around for some
> time. I got this info from the following links -
Good question that has an easy simple answer and a much longer detailed
answer...
Assuming just PC 'desktop' and 'laptop' computers, I'll try a brief stab
at an answer:
Computer electronics are usually designed for 100% utilisation for 100%
of the time. Running your CPU at 100% utilisation is fine.
Newer computers have "green" energy-saving features that are controlled
by "APM" or "ACPM" that can switch off unused devices. Hence, if your
system is idle, it can shut down into an unresponsive 'sleep' state.
Some trigger event then switches things back on to bring the system back
to life. This is very good for laptops to conserve battery power, or
just to reduce generating heat.
A HLT ("Halt") instruction can be used on the CPU which literally stops
all further CPU operation until some trigger event. This reduces the
very large CPU supply current down to just trivial leakage current while
in the halt state. (However, the supply capacitors get hammered by
violent transients when entering or leaving the CPU halt state. This may
reduce their life slightly.)
Conversely, running CPU tasks that heavily utilise the FPU increases
slightly the current consumed over just running integer routines.
So... for running s@h:
The CPU heat output will be slightly higher than for an idle but
otherwise fully powered up system. If your heatsink and fan are ok, then
this is no problem (1 or 2 deg C temperature rise). If you have a
marginal near heat-death system, s@h will be the first application
showing problems or causing problems.
Keeping s@h active means that you have an active system, so no or
limited 'power saving' by ACPM.
On laptops that are a design compromise due to their size and power,
this may reduce their expected life /slightly/. On desktop PCs, there
should be no concern whatsoever.
Further points:
The greatest system stress on computers is during power up and power
down. You could argue that frequent off-on cycling by the 'power-saving'
stuff costs more in wasted time and reduced component life than what is
saved. (In other words, switch your work PC on in the morning, and
switch off just once when you leave work for home.)
Constant CPU load means constant heating which means constant
temperatures which means /minimum/ mechanical stress to give the
greatest lifespan.
There's other factors, negative and positive also.
In summary:
If your machine is switched on doing something, then its spare CPU
cycles may as well get usefully used (for no adverse impact to the
system or you). The extra power consumed is small. 'Wear and tear'
effects due to s@h are negligible amongst all the other effects.
One proviso: Don't have your disks spin-up and spin-down many times an hour!
Note that your computer will become obsolete long before it fails!
Thus is just my opinion. And I have all my machines run at 100% CPU
utilisation while they are powered up. Never burnt out a CPU yet.
(Meanwhile, lots of peripherals have 'come and gone' (:-|)
Happy crunchin',
Martin
--
---------- OS? What's that?!
- Martin - To most people, "Operating System" is unknown & strange.
- 53N 1W - Mandrake 10.0.1 GNU Linux - An OS for Supercomputers & PCs
---------- http://www.mandrakelinux.com/en-gb/concept.php3
Peter Smithson
ml_...@ddnospamddml1dd.co.uk.dd says...
> Note that your computer will become obsolete long before it fails!
Good point but in our cash strapped company, most us are using obsolete
hardware! I've had a fan replaced and someone who sits near me had a
laptop which made a horrible noise all day when the fan bearing went. We
were all happy when the fan finally packed in! Another guy had to sit
his laptop on the air con unit to work it after his fan went! Luckily
they're getting or have new laptops now but you get the idea that fan
wear is a pain on a laptop where I am now. (if you think that's bad then
I'd better not tell you about some of the servers we have!)
I take your point that power up and down is a great wear on components
but I'm not sure that this enegery saving mode actually involves powering
anything off. Also, increased heat increases wear (If I remember my
electronics correctly) - I'll agree that it's fractional and that the
system will be obsolete before it wears out. (but see above!)
Running SETI@home will undo all of AMD's hard work mentioned here -
http://www.amd.com/us-
en/Processors/ProductInformation/0,,30_118_9485_9487^10272,00.html
This applies to desktop computers.
It's not going to stop me running SETI@home but I was interested in
knowing if the FAQ entry was accurate. I think I've got the answer for
Intel & AMD but I've no idea about the other machines.
Cheers
Peter
Martin 53N 1W
> In article <AGQed.361$lv....@newsfe1-win.ntli.net>,
> ml_...@ddnospamddml1dd.co.uk.dd says...
>
>
>>Note that your computer will become obsolete long before it fails!
>
>
> Good point but in our cash strapped company, most us are using obsolete
> hardware! I've had a fan replaced and someone who sits near me had a
> laptop which made a horrible noise all day when the fan bearing went. We
> were all happy when the fan finally packed in! Another guy had to sit
Fans and then HDDs are the most unreliable of the PC hardware, followed
next by PSUs. The standard trick is for a server to have been running
unmolested for a year or two, to be then shut down for an upgrade, to
then find that the fans and a HDD or two don't restart...
[...]
> but I'm not sure that this enegery saving mode actually involves powering
> anything off.
Transistors or some other relay physically switch off the power supply
to various parts, or an equivalent to the CPU halt instruction is issued
to put systems 'to sleep'. Either way, current consumption is stopped
and parts cool down (and spin down).
> Also, increased heat increases wear (If I remember my
> electronics correctly) - I'll agree that it's fractional and that the
> system will be obsolete before it wears out. (but see above!)
The significant aspect of temperature is that increasing temperature
increasingly slows transistor switching. Lifespan is indeed reduced but
this is negligible for normal PC examples (other things will fail before
the electronics).
> Running SETI@home will undo all of AMD's hard work mentioned here -
>
> http://www.amd.com/us-
> en/Processors/ProductInformation/0,,30_118_9485_9487^10272,00.html
If this is about power management, then yes if you keep your system 100%
utilised! Power management (and on demand speed-stepping) is only useful
for laptops. It is also a good marketing gimmick to very _ineffectively_
save the Earth from us.
If you've got the hardware, you get better returns for Mankind and Earth
by thrashing it 100% or more for maximum gain before obsolescence!
(The green power-down stuff is of use to mitigate against those
ignorants too lazy to turn off unused equipment overnight.)
[...]
> It's not going to stop me running SETI@home but I was interested in
> knowing if the FAQ entry was accurate. I think I've got the answer for
> Intel & AMD but I've no idea about the other machines.
This FAQ is a difficult one to answer accurately yet concisely and in an
understandable way... Would you like to script your take on what you've
found?
Regards,
Peter Smithson
ml_...@ddnospamddml1dd.co.uk.dd says...
> This FAQ is a difficult one to answer accurately yet concisely and in an
> understandable way... Would you like to script your take on what you've
> found?
Ah - forcing me to be constructive you fiend.
"Will running SETI@home overload or burn out my CPU?
No. "
How's that? Probably a bit brief.
"Will running SETI@home overload or burn out my CPU?
Not on a normal system, running software should not burn out your
computer - they were designed to run software. Only a faulty computer
would burn out. "
The trouble is, the FAQ is trying to answer a differnet question -
"Will running SETI@home put any additional strain on my computer?"
is what it's answering.
My answer would be -
"There is not significant additional strain in any circumstance.
Traditionaly, computers execute an idle loop to keep them busy all the
time anyway. Computers with power saving features (usually laptops) will
work slightly harder since the power saving features will not work when
running SETI@home."
Yes the answer might cause people a bit more stress but it's more
honest.
Martin 53N 1W
> ml_...@ddnospamddml1dd.co.uk.dd says...
>
>>This FAQ is a difficult one to answer accurately yet concisely and in an
>>understandable way... Would you like to script your take on what you've
>>found?
>
> My answer would be -
>
> "There is not significant additional strain in any circumstance.
> Traditionaly, computers execute an idle loop to keep them busy all the
> time anyway. Computers with power saving features (usually laptops) will
> work slightly harder since the power saving features will not work when
> running SETI@home."
>
> Yes the answer might cause people a bit more stress but it's more
> honest.
Nice summary.
Now there's just the question of how to get it into the FAQ...?
Augustine
37W from 201W when it's running CPDN, according to the UPS load
monitor.
The load can vary between the several BOINC applications, but I would
expect the power used by each of them to be very similar.
Peter Smithson
eva...@mailandnews.com says...
Hmmm - this is the kind of thing that worries me slightly as our UPS
supply is limited (as anyones is). If the UPS doesn't last as long
because I'm running SETI on 6 of our servers then I wouldn't be too
popular! I'm not sure if the hardware/OS I use will have this increase
in power though.
Martin 53N 1W
If you are worried about that, then you should have UPS support software
running that can shut down non-essential services when on battery power,
and then later do a clean shutdown before the battery runs flat.
Also, you have have tested the system and your backups?
Good luck,
David Woolley
"Martin 53N 1W <ml_...@ddnospamddml1dd.co.uk.dd>" <> wrote:
> Peter Smithson wrote:
> > Traditionaly, computers execute an idle loop to keep them busy all the
> > time anyway. Computers with power saving features (usually laptops) will
> > work slightly harder since the power saving features will not work when
> > running SETI@home."
Not really. It's basically still a gross oversimplification.
Of the systems used by most people in recent times it is only
MS-DOS family versions of Windows that have busy idled. Proper
Halt instructions, that cause the CPU to cease accessing memory and
do very little (they don't necessarily dead stop the CPU - the
original purposes were to free memory for DMA and to give a faster
reschedule after an interrupt, although they are now part of power
management), have existed for more than thirty years, to my direct
knowledge. (Definitely true of IBM 360 series[1] and PDP 11s; if
I remember correctly, PDP 7s also had interruptible halt
instructions.)
On Intel Architecture machines, Linux, all versions of Unix and
all NT family Windows issue Halt instructions.
Moreover, thermal loading depends on the instruction mix. A busy
idle is likely to run entirely in L1 cache, so won't heat up
main memory and will use relatively simple paths in the main CPU which
probably consume less power than, say, the FPU. S@H tends to thrash
the cache and parts of the processing will run main memory near flat
out. It also heavily uses floating point. Because of hitting main
memory, the thermal load excess will be more than indicated by the
CPU temperature.
On a machine with no power management (350MHz generation) but running
NT or Linux, there is a several degrees difference, in CPU heatsink
temperature, between idle and S@H. (Caution, the BIOS, in setup mode,
typically doesn't issue Halts - probably because it cannot rely on
interrupts, so the machine will heat up if you try and measure temperature
using the BIOS.)
[1] If I remember correctly, the halt state was set by loading a program
status word, basically the core of the hardware task switching support,
with a specific value, rather than by a dedicated halt instruction.
Peter Smithson
ml_...@ddnospamddml1dd.co.uk.dd says...
> If you are worried about that, then you should have UPS support software
> running that can shut down non-essential services when on battery power,
> and then later do a clean shutdown before the battery runs flat.
Even if we had that software, there's a certain amount of time taken for
machines to shut down. This critical time will be reduced if computers
are using more power.
In our case the "software" is our systems admin bloke running round
shuting machines down which obviously doesn't work too well at weekends!
But - saying the current sytsem isn't great doesn't justify making it
worse.
> Also, you have have tested the system and your backups?
Only in that when I accidentally delete a file our sys admin chap gets
it back from some bank of tapes he has.
Cheers
Peter Smithson
says...
> In article <vX8fd.122$4k7...@newsfe1-gui.ntli.net>,
> "Martin 53N 1W <ml_...@ddnospamddml1dd.co.uk.dd>" <> wrote:
> > Peter Smithson wrote:
>
> > > Traditionaly, computers execute an idle loop to keep them busy all the
> > > time anyway. Computers with power saving features (usually laptops) will
> > > work slightly harder since the power saving features will not work when
> > > running SETI@home."
>
> > Nice summary.
>
> Not really. It's basically still a gross oversimplification.
I think you're missing the purpose of the FAQ. Have a look at the other
answers - they are all very brief and non technical for normal people to
understand. Give your answer to Jo Bloggs and he'll give up after the
first two sentences. Even my programing collegues don't always know
what CPU cache is never mind L1 L2 etc. - they don't need to know.
Never mind a non technical person.
If someone knew enough about computers to understand your answer then I
suspect they would already know the answer!
The question was "Will running SETI@home put any additional strain on my
computer?". Your answer tells us things about L1 cache, PDP 11's and
CPU temprature in BOIS mode but doesn't actually give a yes/maybe/no
answer.
"More" does not equal "Better". The hard thing is to make a short
answer which provides the information normal users will want in an
understandable form. Sure it's not going to be 100% accurate but at the
moment the FAQ entry doesn't even mention the concept of power saving
which I think it should do as it's becoming increasingly popular.
Cheers
Martin 53N 1W
> In article <T1098...@djwhome.demon.co.uk>, da...@djwhome.demon.co.uk
> says...
>
>>
>>>Nice summary.
>>
>>Not really. It's basically still a gross oversimplification.
>
> I think you're missing the purpose of the FAQ. Have a look at the other
> answers - they are all very brief and non technical for normal people to
> understand. Give your answer to Jo Bloggs and he'll give up after the
> first two sentences.
'Twas interesting to me (:-))
A very good answer answering another aspect of the original questions.
> Even my programing collegues don't always know
> what CPU cache is never mind L1 L2 etc. - they don't need to know.
> Never mind a non technical person.
>
> If someone knew enough about computers to understand your answer then I
> suspect they would already know the answer!
True enough.
> The question was "Will running SETI@home put any additional strain on my
> computer?". Your answer tells us things about L1 cache, PDP 11's and
> CPU temprature in BOIS mode but doesn't actually give a yes/maybe/no
> answer.
>
> "More" does not equal "Better". The hard thing is to make a short
> answer which provides the information normal users will want in an
> understandable form. Sure it's not going to be 100% accurate but at the
[...]
The real skill is in answering with 100% accuracy yet with appropriate
detail. This is much better than the patronising 'lies' and lazy 'half
truths' spoon fed to children and spewed out on TV.
The FAQ really needs a multi-level answer for this one (:-P)
...And the answer is "no". The greater details can then be introduced
for 'normal usage', fractionally increased CPU power consumption running
FPU intensive tasks (s@h and others), and power management/saving.
~misfit~
I agree. Emphatically. Otherwise. This current box I'm on runs with a CPU
die temp of around 34°C using office apps/net surfing etc. Start up SETI or
BOINC and that quickly climbs to above 50°C. At a guess I'd say it uses 40
watts more power at least to run SETI and from the above figures you can see
that it certainly generates more heat.
--
~misfit~
~misfit~
> I've done some reasearch on the web and I'm now pretty sure that the
> increased CPU heat only applies to laptops with AMD or Intel chips.
> Something to do with a HLT instruction which has been around for some
> time. I got this info from the following links -
>
> http://grassomusic.de/english/amdk7.htm
> (Details more than anyone could possibly ever want to know about it!)
>
> http://www.tomshardware.com/mobile/20030102/compaq_evo_n1015v-01.html
> (a smaller summary)
>
> http://www.microsoft.com/downloads/details.aspx?FamilyID=2898f8dd-10f8-
> 4107-9f7b-16c5a525de1e&displaylang=en
>
> That last one just explains that the software side of things is
> catered for in Windows XP. I also saw fixes for Linux to cater for
> the HLT instruction.
>
> Maybe laptops with other processors do something similar but I don't
> have access to any of those so I'm not too interested.
>
> So I think the SETI FAQ reply is accurate since most computers are not
> laptops.
Disagree, see my earlier reply in this thread. That's a desktop PC I'm
talking about.
--
~misfit~
Jan Knutar
> Hmmm - this is the kind of thing that worries me slightly as our UPS
> supply is limited (as anyones is). If the UPS doesn't last as long
> because I'm running SETI on 6 of our servers then I wouldn't be too
> popular! I'm not sure if the hardware/OS I use will have this increase
> in power though.
Atleast on *nix, you can tie in events from the UPS straight into the init
process, which is probably the highest authority in your system short of
the kernel, and have it terminate seti immediately on loss of mains, and
restart seti if mains power is restored.
Jan Knutar
> Nice summary.
>
> Now there's just the question of how to get it into the FAQ...?
When you've reached the point of agreeing somewhat in this thread,
you could try formulating a ready faq entry that could be copy/pasted
into the FAQ, and fire it off to Mark Taylor, and perhaps include a
google groups link to this thread so he could get an overview of the
conversation in case he isn't actively following alt.sci.seti anymore.
David Woolley
Martin 53N 1W <ml_...@ddnospamddml1dd.co.uk.dd> wrote:
> > I think you're missing the purpose of the FAQ. Have a look at the other
I wasn't trying to provide content for the FAQ, I was trying to demonstrate
why the proposed answer, and the reasoning behind it, was unsafe.
> ....And the answer is "no". The greater details can then be introduced
The only safe, very simple, answer is "possibly", but that begs the question.
Giving an unqualified "yes" would scare people off. Saying "no" is
misleading.
There are factors that will increase failure rates and factors that
will reduce them. How the machine would be used without S@H is likely
to affect which of these will dominate.
The CPU is unlikely to be the first thing to fail, although the
rate of one failure mode (electromigration) will definitely increase
measurably[1], and S@H makes negligible demand on a disk other than
keeping it spun up.
The original FAQ oversimplifies by ignoring the fact that computers
do have wearout mechanisms.
[1] See, for example, the graph at the end of:
http://www.reliabilityanalysislab.com/tl_hd_0311_PulsedAluminumElectromigration.asp
which suggests that electromigration failure rates will double for about
every 20 to 25 degrees (C) temperature rise.
Martin 53N 1W
> Peter Smithson wrote:
>
[...]
>>
>>I realise it won't do the computers any harm to process data - that's
>>what they were built for. It's just that the FAQ entry suggests they
>
> I agree. Emphatically. Otherwise. This current box I'm on runs with a CPU
> die temp of around 34°C using office apps/net surfing etc. Start up SETI or
> BOINC and that quickly climbs to above 50°C. At a guess I'd say it uses 40
> watts more power at least to run SETI and from the above figures you can see
> that it certainly generates more heat.
34 = Tambient + k * Pnoload
50 = Tambient + k * (Pnoload + 40)
50 - 34 = k * ( ( Pnoload + 40 ) - Pnoload )
16 = k * ( 40 )
k = 16/40 = 0.4 deg C per W
~~~~~~~~~~~~~~~
Are you sure about that +40W?
I don't see anything like that sort of temperature rise. Or is my
heatsink really over x4 better than yours? ( (;-P) )
(For an AMD Athlon 1900+, air cooled.)
Peter Smithson
> Martin 53N 1W <ml_...@ddnospamddml1dd.co.uk.dd> wrote:
> > Peter Smithson wrote:
>
> > > I think you're missing the purpose of the FAQ. Have a look at the other
>
> I wasn't trying to provide content for the FAQ, I was trying to demonstrate
> why the proposed answer, and the reasoning behind it, was unsafe.
Sorry - I suddenly realised that after I posted. But we've already gone
from the complex part. The post before mine was to create a FAQ entry
from a large amount of info.
What was wrong with the FAQ entry - I know it's simplified but what's
inaccurate?
Cheers
~misfit~
> ~misfit~ wrote:
>> Peter Smithson wrote:
>>
> [...]
>>>
>>> I realise it won't do the computers any harm to process data -
>>> that's what they were built for. It's just that the FAQ entry
>>> suggests they
> [...]
>>
>> I agree. Emphatically. Otherwise. This current box I'm on runs with
>> a CPU die temp of around 34°C using office apps/net surfing etc.
>> Start up SETI or BOINC and that quickly climbs to above 50°C. At a
>> guess I'd say it uses 40 watts more power at least to run SETI and
>> from the above figures you can see that it certainly generates more
>> heat.
>
>
> 34 = Tambient + k * Pnoload
> 50 = Tambient + k * (Pnoload + 40)
>
> 50 - 34 = k * ( ( Pnoload + 40 ) - Pnoload )
>
> 16 = k * ( 40 )
>
> k = 16/40 = 0.4 deg C per W
> ~~~~~~~~~~~~~~~
>
>
> Are you sure about that +40W?
>
> I don't see anything like that sort of temperature rise. Or is my
> heatsink really over x4 better than yours? ( (;-P) )
>
> (For an AMD Athlon 1900+, air cooled.)
Not sure about the 40W, was a guess, going by the increase in heat produced.
My CPU is an XP1800 running on a 200Mhz FSB at 2.1Ghz (10.5x) and 1.80vcore
using the AMD-supplied HSF.
--
~misfit~
David Woolley
Peter Smithson <pgsmithson_d...@yahoo.co.uk> wrote:
> What was wrong with the FAQ entry - I know it's simplified but what's
> inaccurate?
It implies that overheating is a threshold effect, when any level of heat,
especially with power applied, will affect the rate of some premature
failures mechanisms. The machine that was running 25 degrees hotter
will fail through electromigration at least twice as fast as when it
is relatively idle. It's likely that it will fail even faster than
that, because the heat sink temperature rise will be less than the core
temperature rise.
Failures are also statistical, so that some machines will fail very
early even though, statistically, the failure may still be very unlikely
in the 5 or so years service life of the machine, and the reduction in
failures due to expansion and contraction might compensate for it in
the overall picture.
Peter Smithson
> Peter Smithson <pgsmithson_d...@yahoo.co.uk> wrote:
>
> > What was wrong with the FAQ entry - I know it's simplified but what's
> > inaccurate?
>
> It implies that overheating is a threshold effect, when any level of heat,
> especially with power applied, will affect the rate of some premature
> failures mechanisms.
I don't mention heat or failure of any components. Just a reminder -
-------
"Will running SETI@home put any additional strain on my computer?"
is what it's answering.
My answer would be -
"There is not significant additional strain in any circumstance.
Traditionaly, computers execute an idle loop to keep them busy all the
time anyway. Computers with power saving features (usually laptops) will
work slightly harder since the power saving features will not work when
running SETI@home."
--------
Perhaps I need to add a sentance saying that the main result is a small
increase in power usage just to avoid misunderstanding it. But even
that isn't really an additional "strain" . I suppose it's a non
technical answer to a non technical question.