CSI stuff:Fact/fiction/both

georges · 9 Aug 2006 at 11:18

Tru said:
As with most productions like this they have technical staff who advise and correct them. They add some pazzazz for effect but generally all the chemical stuff is at least based in reality, chemical names etc will be accurate and I imagine the machines and methods will be similar.

The most outlandish thing I see them do is ZOOM, ENHANCE. They used this technique to get the accurate image of a killer, by zooming and enhancing on the pixels in the eye of someone he photographed.

Yeah.. all american cop shows have that and manage to create an incredibly detailed picture from ~5 pixel

Borris · 9 Aug 2006 at 11:46

aardvark said:
i'm a scientist and so are all my virtual friends

fixed

aardvark · 9 Aug 2006 at 11:54

Borris said:
fixed

damn you borris - if only it wasn't so close to the truth

Borris · 9 Aug 2006 at 12:02

aardvark said:
damn you borris - if only it wasn't so close to the truth

PobodY · 9 Aug 2006 at 12:08

aardvark said:
i'm a scientist and so are virtually all my friends and the only time we watch it is if we need a laugh - the vast majority of stuff shown is laugh out loud wrong - does that answer your question?

I'm an analytical chemist and likewise laugh hard everytime they just "run it through the mass-spec"... without calibration or standards... and it tells them exactly what the unknown is, right down to the only place that sells it.
I asked my boss for one, but he said that AU$250,000 only buys a normal ICP-MS and not a magic one.

I love to watch it though... it drives my wife insane when I criticise them.

Tru · 9 Aug 2006 at 12:22

I just remembered about a massive thread on here that debated why they used IP's that started with 315.xxx.

Beansprout · 9 Aug 2006 at 12:25

Tru said:
I just remembered about a massive thread on here that debated why they used IP's that started with 315.xxx.

Those are the classified military internets

bitslice · 9 Aug 2006 at 12:26

PobodY said:
they just "run it through the mass-spec"... without calibration or standards... and it tells them exactly what the unknown is

heh, nice clean spike too.
not sure what a real lab is like, but we have a whole load of vacuum pumps chugging away - their lab is pretty quiet.

I'd guess in RL they don't turn on the perps PC either, but remove the hard drive instead.

PobodY · 9 Aug 2006 at 12:43

bitslice said:
heh, nice clean spike too.
not sure what a real lab is like, but we have a whole load of vacuum pumps chugging away - their lab is pretty quiet.

I'd guess in RL they don't turn on the perps PC either, but remove the hard drive instead.

*hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm*
(Just pretend it keeps on humming away for the next 23hrs, at a level just below the threshold for OHS regulations)
No chillers either... we have one on each instrument - compressor and a pump running all the time to keep the internals cool.
No warm-up time either... I know that ALL of our plasma instruments take a minimum of 30 minutes to stabalise after the plasma ignights... plus 70 minutes if the optics need to be purged and re-initialised (if the power or argon got interupted). GC would have about the same to get the column to temperature before you could even think about calibrating.

I also like that there is absolutely no accounting for dilution factors either; get a scraping, put it in a tube, add an arbitary amount of solvent (or acid I suppose), shake it, put it on the GC.... and look it was tar from the BP sevrice station on the corner of third and main streets and was laid 1 week ago.
That's one hell of a spectral database they've got... not only common peaks, but propriatary compounds too!

See... now I've got started.

aardvark · 9 Aug 2006 at 13:04

i saw one a few weeks ago when they had some diamonds and then some woman had a handheld scanner and it gave a nice rotating molecule and told them it was made of carbon - i almost fell out of my chair! - it should be called a science fiction show.

Sinizterguy · 9 Aug 2006 at 13:05

Curiousalien said:
Also the chemicals are not that clever and cannot just show up blood they also glow at traces of rust, bleach and poo.

Dont forget copper - the stuff is supersensitive to copper and even trace amounts will show up really well.

CSI do use a lot of factually accurate stuff, some of it they make it look better and easier than it really is to better portray what happens within the timescale they have. After all it is a TV show and needs to keep people watching - if they took as long as they would really take to spray down a place, they wouldnt be able to show anything else.

The most annoying thing for me is how they take a photo that is crap and zoom/enhance it to the point where you can see everything clearly - just not possible.

Meridian · 9 Aug 2006 at 20:59

PobodY said:
I'm an analytical chemist and likewise laugh hard everytime they just "run it through the mass-spec"... without calibration or standards... and it tells them exactly what the unknown is, right down to the only place that sells it.
I asked my boss for one, but he said that AU$250,000 only buys a normal ICP-MS and not a magic one.

I love to watch it though... it drives my wife insane when I criticise them.

As I said earlier, I'm more impressed by the fact that the same mass spec can produce DNA profiles. And even more impressively, it only seems to take about ten minutes.

I remember one episode where Jorja Fox watched someone load a set of DNA samples and then said (about the results): "I'll wait". Bring a book love, 'cos it takes about seven hours...

As for noise levels, the most accurate representation of a forensic lab ever on TV was the first Prime Suspect (the set was a replica of an FSS lab): the camera pans right and there were two scientists gossiping about a third one. Our main lab is pretty quiet (unless one of the girls (I can say that because most of them are young enought to be my daughters) decides to put the radio on and won't take "turn that off or I'll make you eat it" for a no), but the instrument is pretty load - mostly from cooling fans and the oven fan on the GC/MS.

Standards...

That one is actually a bit complicated: it depends on what you are doing. Yes, if you are measuring a purity then standards will always be run. Pure analytical stuff is different: if you have a library you trust then you can take the results from that and accept them if you wish - even UKAS accept this. If you're not familiar with the compound of interest then standards it is. If you don't trust the library then it's standards again - or backup from a text-book you trust. For instance, one commercial IR library from an American forensic lab contains at least two wrong spectra.

aardvark

As I said earlier, actually most of of the science is correct in theory - it's just the practice that's wrong. And the effacacy of most of the techniques is greatly exaggerated. The various bits of kit all work, it's just that they are aren't as good as is shown. Very few things on forensic science are definite - even DNA (famously) just quotes odds.

M

aardvark · 9 Aug 2006 at 21:26

the analytical methods maybe based on fact but in the very few episodes i've seen (cos it annoys me so much - my friend watches it and tells me about it sometimes) there have been lots of really simple chemistry errors.

alexakasloth · 9 Aug 2006 at 21:51

I'm watching an ep from season 1, so quite a few years ago.
"You guys get some toys"
"It's not a toy O'Reilly, it's an electro static dust print lifter"

O Rly?

Meridian · 9 Aug 2006 at 22:05

aardvark said:
the analytical methods maybe based on fact but in the very few episodes i've seen (cos it annoys me so much - my friend watches it and tells me about it sometimes) there have been lots of really simple chemistry errors.

It's usually more a case of them just making up some BS because it would a) take too long to explain the real chemistry, and b) the real chemistry wouldn't actually be relevant to whatever it is they are trying to prove. CSI:Miami was the worst offender here with an episode in the first serious about some new drug in which the chemical diagram they drew made no sense at all.

M

aardvark · 9 Aug 2006 at 22:27

Meridian said:
It's usually more a case of them just making up some BS because it would a) take too long to explain the real chemistry, and b) the real chemistry wouldn't actually be relevant to whatever it is they are trying to prove. CSI:Miami was the worst offender here with an episode in the first serious about some new drug in which the chemical diagram they drew made no sense at all.

M

that might be the episode my friend told me about - he said it was pretty funny.

i remember one where they found a rag that had been outside for a few hours - one person said 'that smells like alcohol' and the other confirmed that it was ether - for one thing ether smells nothing like alcohol and if it was out side for more than 10 minutes all the ether would have evaporated anyway and you wouldn't smell a thing - little things but easy to fix if they actually employed an expert to help their scripts - or maybe i'm just too picky

Brizzles · 9 Aug 2006 at 22:41

I must admit i laugh at how they enlarge a blurred reflection off a mosquito's contact lens from half a mile into a passport photo of the suspect.

Even with my limited knowledge of digital photography and photoshop this isn't possible. Unless every security camera in America is 40 million mega pixel.

But , it's a fictional drama , and a good one IMO , so cut it some scientific slack.

If we wanted true to life , then , as someone posted before , waiting 7 hours for a DNA / Mass spectrometer result might not be everyones ideal veiwing.

Cheers,

Mark

Meridian · 10 Aug 2006 at 07:27

mrdbristol said:
But , it's a fictional drama , and a good one IMO , so cut it some scientific slack.

I do, and I watch all the series (even if David Caruso annoys the **** out of me) and enjoy them. But just as people here will mock when there are bad computer bits on TV/film, so we will mock when we see dubious forensic science. But the series are still a mile ahead of Waking the Dead, which features some shockingly bad forensic science.

M

PobodY · 10 Aug 2006 at 11:08

aardvark said:
i saw one a few weeks ago when they had some diamonds and then some woman had a handheld scanner and it gave a nice rotating molecule and told them it was made of carbon - i almost fell out of my chair! - it should be called a science fiction show.

That sounds like a portable XRF - real technology, just not that accurate nor do they give you 3D images.
They're useful for geologists to get a rough idea before they send it in to get assayed. You have to pick the source etc you want before hand, so reall need to have a good idea about what you expect to read with it.

I'm sorry Meridian, I come from an inorganic viewpoint - it's ALL about standards for us. Esepecially if you're using an ICP-MS to finger-print something (like gold bullion for example).

Meridian · 10 Aug 2006 at 17:26

PobodY said:
I'm sorry Meridian, I come from an inorganic viewpoint - it's ALL about standards for us. Esepecially if you're using an ICP-MS to finger-print something (like gold bullion for example).

That's fair enough, because it's your equivalent of a quant anyway. But our feeling is that for normal benchtop quadropole, once a validated standard has been run, that's it. As long as the tune is within parameters and as long as the GC test mix works (along with the RT lock) then we are good to go for simple analysis - we don't need to keep running the same standard. But if it throws up something where we haven't run a validated standard then we may try to obtain some - depends on the case and the importance of the evidence.

M