G45 GBase Thinktank

I tweaked the auto-suggest for a few hours, and added the "2-sider" option, which simply adds 15% to the total value. The result is also adjustable for inflation via a fractional multiplier. Im confident now that the computed values are useful and mostly compare well with real world results. Actually the records I tested yielded results amazingly close to what I would expect, and to Popsike. The function of the "accept" button has also been improved.

http://www.g45central.com/Gtest/nutest4.html

(in purple Collector/G45 section).
 
An interesting tool to fool around with, but just open to someone else opinion to abuse or misuse like any other price guide. Just a bit more complicated. Like TBM ratings one mans 5 is another mans 1. I did not notice an adjustment for quantity finds on the final price either. Unless the numbers used for computation will be within the data base for each record valued after this gets off the ground. I do not see how you can get a usable value without using the Band , record label, etc. too.
 
vinyl01 said:
An interesting tool to fool around with, but just open to someone else opinion to abuse or misuse like any other price guide. Just a bit more complicated. Like TBM ratings one mans 5 is another mans 1. I did not notice an adjustment for quantity finds on the final price either. Unless the numbers used for computation will be within the data base for each record valued after this gets off the ground. I do not see how you can get a usable value without using the Band , record label, etc. too.
Click to expand...

Yes.
However, at least with the Auto-suggest tool, there is some understandable, consistent logic behind each valuation. Unlike a price guide which gives you nothing but an opinion. Plus you can input your own values to check the result validity. For example, if you don't agree with the Song or Legend rating, just dial in your own value and check the result.

The band is taken care of by the Legend factor. The Record label (visual) can be incorporated into the Artifact factor or (non-visual) the Legend factor. Quantity Finds are handled by the Rarity factor. The S/R/L/T/A system is very comprehensive, covering almost every collectible aspect of a 45.

The computation is quite complicated, so the numbers themselves won't help most users. Here they are:

function autosuggest()
{
var s=document.getElementById("g45s").value;
var r=document.getElementById("g45r").value;
var l=document.getElementById("g45l").value;
var t=document.getElementById("g45t").value;
var a=document.getElementById("g45a").value;
var d1=document.getElementById("def1");
d1=d1.options[d1.selectedIndex].value;
var dl1=document.getElementById("deflev1");
dl1=dl1.options[dl1.selectedIndex].value;
var d2=document.getElementById("def2");
d2=d2.options[d2.selectedIndex].value;
var dl2=document.getElementById("deflev2");
dl2=dl2.options[dl2.selectedIndex].value;
var w=document.getElementById("twosider").checked;
var wfac=1;
if (w==true)
{wfac=1.15}

s=parseInt(s);
r=parseInt(r);
l=parseInt(l);
t=parseInt(t);
a=parseInt(a);
d1=parseInt(d1);
dl1=parseInt(dl1);
d2=parseInt(d2);
dl2=parseInt(dl2);

var d=document.getElementById("def1");
var df=d.options[d.selectedIndex].text;
if (df=="Off-Center Label" || df=="Labels Reversed" || df=="Label Misprint" || df=="Hole Drilled")
{
var d1fac=d1;
}
else
{
var d1fac=(d1*dl1)/10;
}

var d=document.getElementById("def2");
var df=d.options[d.selectedIndex].text;
if (df=="Off-Center Label" || df=="Labels Reversed" || df=="Label Misprint" || df=="Hole Drilled")
{
var d2fac=d2;
}
else
{
var d2fac=(d2*dl2)/10;
}

var dfac=(d1fac+d2fac);

if (s !==0 && r !==0 && l!==0 && t!==0 && a !==0)
{

dfac=100-dfac;
if (dfac<=0){dfac=1;}

var gfac=.9756;

var sp=s/1.9;
var rp=r;
var lp=l/1.5;
var tp=t/1.6;
var ap=a/1.7;

var sprp=(sp+rp)/1.80;
var sprpx=Math.exp(sprp);
var lp=lp/1.45;
var lpx=Math.exp(lp);
var tp=tp/1.45;
var tpx=Math.exp(tp);
var ap=ap/1.55;
var apx=Math.exp(ap);

var dv=(s+r)/4;

var totfacs = (sprpx/1.6) + lpx*dv + tpx*dv + apx*dv;
xval=((totfacs*gfac*wfac)/100)*dfac;

xval=Math.floor(xval);
document.getElementById("suggested").value=xval;
}
}
 
MopTopMike said:
Quite intriguing! But what about a mint minus 45 with some sort of manufacturing defect, like reversed labels, or bumps in the wax?
Click to expand...
That would not be difficult to program, but it could be quite hard to quantify for each different type of defect. Some wax bubbles are hardly noticeable, others completely ruin a disc. Same with the other defects. But if we can come up with an agreed table of defects and relative negative scores associated with each defect, then I could easily incorporate those numbers into the Auto-suggest function and interface.
 
I surmise that some defects would need to be quantified by a percentage - say, 75% of the label has wear, or there is a sticker / label tear covering 1/3 of the label but the wax itself is near mint... or, the the surface of the wax has 20% bubbles in the wax.
For serious defects, such as warps, and off-center pressings which usually impede enjoyment of playback, those might need to be fixed and not variable. Finding a non off-center "Come On Mary" by the Abandoned would be a difficult and possibly futile search.
 
I have made the fine adjustments. Below is the list of weighting factors for each defect. Does anyone (dare to :sneaky:) disagree?

No Defects : 0% reduction
Pressing Flaw : 1 = 10% reduction, 10 (unplayable and unfixable)=99% reduction
Off-Center Pressing : 1 = 5% reduction, 10 = 50% reduction
Off-Center Label : flat 15% reduction
Label Misprint : flat 3% reduction
Labels Reversed : flat 7% reduction
Water Damage : 1=6% reduction, 10 (bad stains on wax and label)=60% reduction
Torn Label : 1=4% reduction, 10=40% reduction
Writing On Label : 1=3% reduction, 10=30% reduction
Sticker On Label : 1=2% reduction, 10=20% reduction
Warp : 1=10% reduction, 10 (unplayable and unfixable)=99% reduction
Hazy Wax : 1= 2% reduction, 10 = 20% reduction
Mouldy Wax : 1= 7% reduction, 10=70% reduction
Bubbly Wax : 1= 5% reduction, 10=50% reduction
Cracked Wax : 1= 10% reduction, 10 (unplayable) = 99% reduction
Hole Drilled : flat 5% reduction
Broken Wax : 1 = 10% reduction, 10 (unplayable)=99% reduction

http://www.g45central.com/Gtest/nutest4.html

I also moved the repro and quantity info into the yellow "Pressing" page, to avoid confusion in the purple "G45 Ratings & Values" page
 
Disagree! Hazy wax, never buy them! Who knows what The Haze will lead to 10 years from now. I think a maximum of 20% should only be the max for very rare discs. A hazy $500 disc is not welcome in my home (uless it has hazed in my bunker...)
 
Mr.Segment said:
Disagree! Hazy wax, never buy them! Who knows what The Haze will lead to 10 years from now. I think a maximum of 20% should only be the max for very rare discs. A hazy $500 disc is not welcome in my home (uless it has hazed in my bunker...)
Click to expand...

Calling all collectors - do you agree with this? Personally light haze does not worry me too much, but heavy or patchy haze is a worry.

Should the haze defect cause a bigger price reduction than 20% max?
Maybe Level 1 should be 5%, Level 10 should be 50%....?
 
bosshoss said:
I have made the fine adjustments. Below is the list of weighting factors for each defect. Does anyone (dare to :sneaky:) disagree?

No Defects : 0% reduction
Pressing Flaw : 1 = 10% reduction, 10 (unplayable and unfixable)=99% reduction
Off-Center Pressing : 1 = 5% reduction, 10 = 50% reduction
Off-Center Label : flat 15% reduction
Label Misprint : flat 3% reduction
Labels Reversed : flat 7% reduction
Water Damage : 1=6% reduction, 10 (bad stains on wax and label)=60% reduction
Torn Label : 1=4% reduction, 10=40% reduction
Writing On Label : 1=3% reduction, 10=30% reduction
Sticker On Label : 1=2% reduction, 10=20% reduction
Warp : 1=10% reduction, 10 (unplayable and unfixable)=99% reduction
Hazy Wax : 1= 2% reduction, 10 = 20% reduction
Mouldy Wax : 1= 7% reduction, 10=70% reduction
Bubbly Wax : 1= 5% reduction, 10=50% reduction
Cracked Wax : 1= 10% reduction, 10 (unplayable) = 99% reduction
Hole Drilled : flat 5% reduction
Broken Wax : 1 = 10% reduction, 10 (unplayable)=99% reduction

http://www.g45central.com/Gtest/nutest4.html

I also moved the repro and quantity info into the yellow "Pressing" page, to avoid confusion in the purple "G45 Ratings & Values" page
Click to expand...

Good enough. When do you think this will be up and running to start inputting our data?
 
bosshoss said:
I have made the fine adjustments. Below is the list of weighting factors for each defect. Does anyone (dare to :sneaky:) disagree?

No Defects : 0% reduction
Pressing Flaw : 1 = 10% reduction, 10 (unplayable and unfixable)=99% reduction
Off-Center Pressing : 1 = 5% reduction, 10 = 50% reduction
Off-Center Label : flat 15% reduction
Label Misprint : flat 3% reduction
Labels Reversed : flat 7% reduction
Water Damage : 1=6% reduction, 10 (bad stains on wax and label)=60% reduction
Torn Label : 1=4% reduction, 10=40% reduction
Writing On Label : 1=3% reduction, 10=30% reduction
Sticker On Label : 1=2% reduction, 10=20% reduction
Warp : 1=10% reduction, 10 (unplayable and unfixable)=99% reduction
Hazy Wax : 1= 2% reduction, 10 = 20% reduction
Mouldy Wax : 1= 7% reduction, 10=70% reduction
Bubbly Wax : 1= 5% reduction, 10=50% reduction
Cracked Wax : 1= 10% reduction, 10 (unplayable) = 99% reduction
Hole Drilled : flat 5% reduction
Broken Wax : 1 = 10% reduction, 10 (unplayable)=99% reduction

http://www.g45central.com/Gtest/nutest4.html

I also moved the repro and quantity info into the yellow "Pressing" page, to avoid confusion in the purple "G45 Ratings & Values" page
Click to expand...

Are you not just fine tuning the M- / G- grading here and creating an unrealistic equation of impossibility and confusion? Over egging the cake comes to mind:p.

..and I have other faults on a few 45's here if you are going down this path - labels unstuck from disc but exist in a bag, Label unstuck and lost forever, Foreign object attached to disc - e.g. paint splashes / glue / snot / food?

And have % additions for things like - promo photo included + 20%, Autographed by band member + 15%, Song sheet included +10% etc....:D
 
There are probably too many variables that might be missed to program them all. Even ones with drill holes, labels off center, degrees of warps, amount of wol, stol, tol, etc., etc, etc.
 
vinyl01 said:
Good enough. When do you think this will be up and running to start inputting our data?
Click to expand...

I can only guess, but let's say 30 days from now? It will have to be tested via a small panel before it's made public. The last thing I want is 50 people all complaining at once...:nope:... 5 at once would be enough...:yup:
 
bosshoss said:
Calling all collectors - do you agree with this? Personally light haze does not worry me too much, but heavy or patchy haze is a worry.

Should the haze defect cause a bigger price reduction than 20% max?
Maybe Level 1 should be 5%, Level 10 should be 50%....?
Click to expand...

I agree with you, Mark. A light haze is no big deal. That rarely affects the audio quality.

Now, Buffed 45s is a different matter.

The only percentage I find fault with is the reversed label pressing error. My personal most hated problem with 45s. I'd rather have a cracked but plays perfect (no noise) 45 than one with flip-flopped labels. I'd make that a 25% less factor in the algorithm.
 
vinyl01 said:
There are probably too many variables that might be missed to program them all. Even ones with drill holes, labels off center, degrees of warps, amount of wol, stol, tol, etc., etc, etc.
Click to expand...

It is a bit confusing the way I've presented it, but...

The defect checkboxes in the yellow "pressing" section are the ones that will be saved for the database. They represent faults noted across the entire pressing, or a large number of the known copies. The defect drop-downs in the purple section are really for valuing individual specimens. But I left them the Console because they will be useful when valuing records for the database. I will definitely put the Autosuggest Valuation module up on G45central as a separate entity, at a later date.
 
Hicksville Hogwash said:
Are you not just fine tuning the M- / G- grading here and creating an unrealistic equation of impossibility and confusion? Over egging the cake comes to mind:p
Click to expand...

Thanks Hogwash! This cake is going to be the most over-egged one ever baked :tiphat:

..and I have other faults on a few 45's here if you are going down this path - labels unstuck from disc but exist in a bag, Label unstuck and lost forever, Foreign object attached to disc - e.g. paint splashes / glue / snot / food?
Click to expand...

I think these would fit into the existing framework
- labels unstuck="pressing flaw, level 5", gone forever="Level 7"
- paint splashes, glue snot and food are all part of the actual record grading..ie vg minus or g

And have % additions for things like - promo photo included + 20%, Autographed by band member + 15%, Song sheet included +10% etc....:D
Click to expand...

That's a great idea! But it's already sort of accounted for with the Artifact rating. I'll think about it, though.

After thinking a little - I think the Artifact rating should be weighted more heavily in the 5-10 range, perhaps up to 20% addition, to allow for the inclusion of inserts, autographs, photos, etc. etc. These variations are too numerous and diverse to be given individual checkboxes and values. However, in the AutoSuggest 1.0 Module Instruction Manual that will be shipped with the final product, I will list the Artifact ratings appropriate to use for all the various additions. The lower ratings (1-4) will still be appropriate for the original purpose of noting cool label design etc.
 
You must log in or register to reply here.

Similar threads

Al.Burnett1966
The Great Wall 4-song acetate
Replies
10
Views
405
The Raven
The Raven
Al.Burnett1966
60's Forgotten Things Compilation LP - Welcome To The Wilderness
Replies
6
Views
766
Al.Burnett1966
Al.Burnett1966
M
Love Me (Don't) ?
Replies
2
Views
230
masterbeat64
M
M
Living without the Past
Replies
7
Views
530
masterbeat64
M
chas_kit
Monarch, Alco, AFM Engineering
Replies
2
Views
324
chas_kit
chas_kit