Quantifying opacity in TNs

Anything to do with Port.
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Quantifying opacity in TNs

Post by DRT »

Split by DRT from the Sandeman 1977 tasting note.
In a [url=http://www.theportforum.com/viewtopic.php?f=2&t=5593&p=47399#p47399]TN for S77[/url], jdaw1 wrote:
jdaw1 wrote:S77 ! 40% opaque.
AHB wrote:80% opaque.
That is a massive discrepancy.

Other discrepancies in F77, Ni77, Mg77, Cá77.
At what time in the evening were the notes taken? Is it possible that JDAW noted the early colour and AHB, many hours later, noted a darkening in the glass/decanter through longer exposure to air?

Was JDAW closer to a light source than AHB?
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: 1977 Sandeman

Post by DRT »

Whilst I was posting the above JDAW has been posting lots of similar notes of discrepancy in the related TN threads.

Conclusion: Either AHB requires glasses or JDAW has X-ray vision. I really hope is the latter as that would make our tasting group quite unique and would solve the difficulty of checking levels in black bottles.
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: 1977 Sandeman

Post by jdaw1 »

(And also editing the earlier post in this thread to link to those discrepancies.)

But both AHB and I thought that the Df77 was 40% opaque.
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: 1977 Sandeman

Post by DRT »

jdaw1 wrote:(And also editing the earlier post in this thread to link to those discrepancies.)

But both AHB and I thought that the Df77 was 40% opaque.
But did you have your X-ray filter turned on at that point in the evening?
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: 1977 Sandeman

Post by jdaw1 »

Another line of reasoning. Somebody who wasn’t there: what opacity would you expect of a mixed bunch of 1977s? Top-quartile? Bottom quartile? Or inter-quartile (25% to 75%)?

Next, look at the pictures in the review thread. What might those opacities be? Notice the Ni77 in four pictures.

DRT wrote:But did you have your X-ray filter turned on at that point in the evening?
Shhhh: the earthlings don’t know.
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: 1977 Sandeman

Post by DRT »

Based on the first picture in the review thread I would rate the opacity of these wines as:

GC 100%
Cá 80%
T 70%
F 60%
Ni 40%
Df 40%
S 40%
Mg 30%

I have not compared that list to JDAW & AHB's assesments.

As for the 1980 vintage in general, I would expect wines of that age to be somewhere between what we see for Fonseca and Niepoort in this line-up. I think the three at the top are exceptionally dark for a 32 year old VP, the top two perhaps inexplicably so. In this line-up only the Morgan looks like it has matured too quickly.
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
JacobH
Quinta do Vesuvio 1994
Posts: 3300
Joined: 16:37 Sat 03 May 2008
Location: London, UK
Contact:

Re: Quantifying opacity in TNs

Post by JacobH »

I have never quite understood what opacity refers to when expressed as a percentage in a tasting note. Within the context of a glass-sized sample, isn’t a Port either opaque or transparent? I understand a lighter transparent Port lets through more light than a darker one, but is that not reflected in colour since the two strongly collateral?

(I post this out of genuine confusion and interest rather than disagreement!)
Image
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: Quantifying opacity in TNs

Post by DRT »

JacobH wrote:I have never quite understood what opacity refers to when expressed as a percentage in a tasting note. Within the context of a glass-sized sample, isn’t a Port either opaque or transparent? I understand a lighter transparent Port lets through more light than a darker one, but is that not reflected in colour since the two strongly collateral?

(I post this out of genuine confusion and interest rather than disagreement!)
Would 60% Co-paque make more sense? :lol:
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
WS1
Cruz 1989
Posts: 1058
Joined: 23:08 Wed 04 Feb 2009
Location: Hertfordshire

Re: Quantifying opacity in TNs

Post by WS1 »

Hi there,

generally I agree with Derek's coments and rating on the 1977 ports we had. When receiving Julian's reminder about TNs and saw the discussion I was very tempted to post my opinion but realised I had not finished yet the TNs of our gathering on the 20th of December last year yet as I promised to do. In fearing backlash from Julian I stayed quiet.
After this thread was kindly opend in the conversation section I can easily post my opinion. :D
From memory (I did not take tasting notes that night) the colour of the Sandeman was very bright and flashy. This made the wine look darker. Also the dense of the colour (if you attempt to look through the port was better than the similarily coloured Delaforce. Also the colour of Delaforce lacked the brightness the Sandemann had. As a comparison the colour of Calem was a lot darker but the composition and density is nowhere near as good as on the Fonseca and Taylor.
The discrepancy in the "opacity" imho is due to different definitions/understanding of the individual taster. Personally i divide up the appreance of a wine/port in at least into three sections:

- pure colour
- density/composition of the colour (brightness, can you see through the wine, how quickly the rim becomes watery, how quickly the colour gets lighter towards the rim etc. )
- how quickly the port wanders down after rotating the glass

regards

WS1
"Sometimes too much to drink is barely enough"
Mark Twain
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: 1977 Sandeman

Post by jdaw1 »

DRT wrote:As for the 1980 vintage in general
This was a tasting of 1977s.
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: 1977 Sandeman

Post by DRT »

jdaw1 wrote:
DRT wrote:As for the 1980 vintage in general
This was a tasting of 1977s.
Sorry, I had only just finished typing something about 1980 when I wrote the above summary. I did mean 1977, and should have said 35 year old rather than 32, but allowed my arithmetic to compound the original error.
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

JacobH wrote:I have never quite understood what opacity refers to when expressed as a percentage in a tasting note.
Visit a church. Look at the stained-glass windows. Find one that is blue, but lets through about 20% of the light. I call that ‟blue, 80% opaque”. There might be other windows, of a similar hue (a similar version of blue, perhaps â—¼ rather than â—¼ or â—¼), but different opacity.

Likewise with port. There is a glorious red colour (on my screen resembling â—¼) that comes in different opacities. Some might prefer to think of them as different densities.

To assess, look at the circle text through about 1cm of port in a titled glass.

And it is my only contribution to TN terminology, having been adopted by AHB. :-)
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: Quantifying opacity in TNs

Post by DRT »

jdaw1 wrote:To assess, look at the circle text through about 1cm of port in a titled glass.
I think we might have found the source of the discrepancy.

I have, very infrequently, noted the opacity of ports at tastings. But I do so by looking straight down through the centre of the glass before I drink from it. My method would result in a higher % than Julian's method in most instances. I do not know which method AHB uses.
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

DRT wrote:I do not know which method AHB uses.
The eye compensates fairly well for the view angle. And, usually AHB and I are within ±10% of each other.
Glenn E.
Graham’s 1977
Posts: 4172
Joined: 22:27 Wed 09 Jul 2008
Location: Seattle, WA, USA

Re: Quantifying opacity in TNs

Post by Glenn E. »

I don't normally rate opacity, with the main exception being for very young Ports.

When rating very young Ports, I hold a standard size pen or pencil behind the glass (which is held at a 45 degree angle) right at the edge of where the Port meets the glass, and then judge how much of the pencil's width is obscured by opaqueness. (Is that a word?) In very young Ports it is normal for at least part of the pencil to be obscured, but I have seen a few where the entire width of the pencil was barely visible.

Also occasionally I'll note that a Port (of any age) is opaque in the center, which simple means that you can't see through the center of the glass.

The reason that I don't use a percentage is that I really don't know how to judge light transmission through an object. I have tinted windows on my car that I had installed by a professional, but I cannot figure out how they arrive at the "35%" number that's supposed to tell me how dark my windows are. (That may or may not be the actual number... I can't remember what it was. Mine are as dark as is allowed by law in Washington, which really isn't that dark.) My assumption is that water or clear glass would be 0%, but as I recall that didn't fit with the numbers used for window tinting.

Since I can't even figure it out on window tinting, which is well established and regulated by law, I can't expect to come up with useful numbers for Port where there might be several factors conspiring to change my perception.
jdaw1 wrote:Somebody who wasn’t there: what opacity would you expect of a mixed bunch of 1977s? Top-quartile? Bottom quartile? Or inter-quartile (25% to 75%)?
Assuming that higher opacity means a darker Port, I would expect a mixed bunch of 1977s to range from no worse than 2nd quartile all the way up into the top quartile with most of them falling in the 3rd quartile. This is based partially on my own experiences with 1977s, but mostly on the reputation of the vintage as a whole. IIRC the GC77 that I had would probably fit into the top quartile, as it was quite dark and very nearly opaque in the center. I've had a D77 and an S77 that would probably both fit in the 3rd quartile. Off the top of my head those are the only ones I can remember right now.
Glenn Elliott
PhilW
Dalva Golden White Colheita 1952
Posts: 3503
Joined: 14:22 Wed 15 Dec 2010
Location: Near Cambridge, UK

Re: Quantifying opacity in TNs

Post by PhilW »

My understanding of opacity would be the proportion of light transmitted compared to the source; this therefore depends on the illumination source, the frequency of the light, the medium through which the light is transmitted, the distance through the medium and the frequency response of the detector.

For our purposes with port, if the eye is assumed to be a linear detector (I do not know if this is valid) and we are not close to the noise limit due to insufficient or too much light, then the most likely cause of error would be variation in the width of the port through which the light must pass. Experimentation would likely be required to determine likely error; while it could probably be minimised within glasses at a single tasting, keeping this measure valid for comparison between tasting would be difficult, unless some means of standard depth could be devised (such as using a tasting bottle with a specific fill level).

The suggestion of then using a sheet with 100%/80%/60% etc marked would then also rely on a standard calibrated light source; even if the response of the eye were fixed, which it is not since it is adaptive which complicates further. One solution would be a calibrated measure - effectively a series of known opacity slides - which could then be used for comparison against a pre-printed sheet with varying density print; this would also remove the need for the print density itself to be calibrated or sheet whiteness etc to be an issue.
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: Quantifying opacity in TNs

Post by DRT »

Isn't it wonderful how a seemingly simple requirement can be turned into an unfathomable problem when the IT guys get involved :lol:

I feel a Dilbert coming on!
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

PhilW: the eye works well over several orders of magnitude of illumination. It automatically, even invisibility, compensates for ambient conditions. This is why AHB and I, with very little prior standardisation, are usually so close in our estimates. Might this automatic compensation eliminate some of your concerns?
PhilW
Dalva Golden White Colheita 1952
Posts: 3503
Joined: 14:22 Wed 15 Dec 2010
Location: Near Cambridge, UK

Re: Quantifying opacity in TNs

Post by PhilW »

jdaw1 wrote:PhilW: the eye works well over several orders of magnitude of illumination. It automatically, even invisibility, compensates for ambient conditions. This is why AHB and I, with very little prior standardisation, are usually so close in our estimates. Might this automatic compensation eliminate some of your concerns?
"even invisibility" - what kind of language is that?! :shock:

To answer the specific point, yes it does help reduce any potential overall error, but I am uncertain to what degree. I expect the estimates might be quite different made in a room lit by candles compared with flourescent lights; of course it may be that the typical rooms we eat in fall in a narrow enough suitable range that this issue is minimal.
DRT wrote:Isn't it wonderful how a seemingly simple requirement can be turned into an unfathomable problem when the IT guys get involved :lol:

I feel a Dilbert coming on!
Would it be one where management completely fail to specify their full requirements, blame someone else when the resultant product isn't up to scratch and then fire the engineers who point out the failings and ways to solve it? :D
Glenn E.
Graham’s 1977
Posts: 4172
Joined: 22:27 Wed 09 Jul 2008
Location: Seattle, WA, USA

Re: Quantifying opacity in TNs

Post by Glenn E. »

PhilW wrote:To answer the specific point, yes it does help reduce any potential overall error, but I am uncertain to what degree. I expect the estimates might be quite different made in a room lit by candles compared with flourescent lights; of course it may be that the typical rooms we eat in fall in a narrow enough suitable range that this issue is minimal.
I think that's at least partly Julian's point - the human eye (or rather, the brain) is an excellent compensator and it does it automatically. Provided you have a standard to judge by, humans do a very good job of adjusting for ambient differences. A marked gradient printed on the placemats would simple make this even easier for most to do.
PhilW wrote:Would it be one where management completely fail to specify their full requirements, blame someone else when the resultant product isn't up to scratch and then fire the engineers who point out the failings and ways to solve it? :D
That reminds me of a joke.

A man is floating along in a hot air balloon and sees another man on the ground below. He calls out and asks, "Excuse me, but can you tell me where I am? I seem to be lost." The man on the ground replies, "You're in a hot air balloon."

The man in the balloon thinks for a moment and says, "You must be an engineer."

Surprised by this, the man on the ground replies "Why yes, I am, but how did you know?"

"Well," says the man in the balloon, "while technically correct your answer was completely useless and did not help me with my problem."

The man on the ground paused to think for a second, then replied, "In that case you must be a manager."

Stunned, the man in the balloon replied, "That's correct, but how could you have possibly figured that out?"

"It's very simple," the man on the ground replied. "You had a problem before you met me, we talked briefly, and you still have the exact same problem, yet somehow now it is all my fault."
Glenn Elliott
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: Quantifying opacity in TNs

Post by DRT »

PhilW wrote:
DRT wrote:Isn't it wonderful how a seemingly simple requirement can be turned into an unfathomable problem when the IT guys get involved :lol:

I feel a Dilbert coming on!
Would it be one where management completely fail to specify their full requirements, blame someone else when the resultant product isn't up to scratch and then fire the engineers who point out the failings and ways to solve it? :D
No, not that one. It's the one when the user specifies exactly what he wants and the engineer has to say it's a rubbish idea because he didn't think of it first. :wink:
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
DRT
Fonseca 1966
Posts: 15779
Joined: 23:51 Wed 20 Jun 2007
Location: Chesterfield, UK
Contact:

Re: Quantifying opacity in TNs

Post by DRT »

Glenn E. wrote:"It's very simple," the man on the ground replied. "You had a problem before you met me, we talked briefly, and you still have the exact same problem, yet somehow now it is all my fault."
Typical. As soon as you point out to an engineer that he is being unhelpful he gets moody and defensive.
"The first duty of Port is to be red"
Ernest H. Cockburn
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

[url=http://www.theportforum.com/viewtopic.php?p=48465#p48465]Here[/url], in mid-Feb 2012, jdaw1 wrote:We tried the opacity test, from two different laser printers. The printers were too different, but both useless. Good experiment; bad result. Scrapped.
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

Just added to the log page:

Image

(That text, to facilitate searching: “Opacity comparator, trial version: comments to www.ThePortForum.com/viewtopic.php?t=5610&view=unread#unread”.)

◊ Which way should the shading go? Left = blacker, or left = whiter?

◊ Which way should the numbers go? Higher = blacker, or higher = whiter?

PDFs currently containing this ‘Opacity comparator’:
08 June 2017, 1967s;
10 Oct 2017, Gould Campbell;
30 Nov 2017, 15th anniversary of the software.
User avatar
jdaw1
Cockburn 1851
Posts: 23613
Joined: 15:03 Thu 21 Jun 2007
Location: London
Contact:

Re: Quantifying opacity in TNs

Post by jdaw1 »

Update:
Image

Opacity, in tens of percentage points, indicated by the point at which the white waves become invisible. Hopefully.

It adds 739 bytes to the PDF.
Post Reply