open All Channels
seplocked Science and Industry
blankseplocked Jita Profitability Index
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 2 3 4 [5]

Author Topic

Dr BattleSmith
PAX Interstellar Services
Posted - 2010.03.11 00:46:00 - [121]
 

Originally by: Xenofur

A long time actually, long enough to popularize it. And only within The Forge. And I'm betting on the group of Eve players at large providing enough people who're lazy enough to not care if it might not return the exact profits.


Lets assume there are lots of people using it.....

How long do you expect individuals to throw isk into product-lines which are flooded before realising the tool is burning their wallets?


Sure there will be fresh morons along each day who believe they can pick the top item and be onto instant riches.

Once they waste their isk producing something that isn't selling, has a shrinking margin, and way too much volume. What do they do next?


You claim this tool will destroy Jita but the truth is the tool destroys the profit on any item it lists as being high profit for a short period.
The product-lines aren't always mainstream and the majority of the people building them would be users of the tool.
We can safely conclude that the only people the tool will hurt are the users of the tool itself.

Using it blindly is a recipe for producing exactly the *wrong* items. Anything listed as high-profit is in reality flooded and low-profit by the time production jobs finish.

Xenofur
Aliastra
Posted - 2010.03.11 00:55:00 - [122]
 

Edited by: Xenofur on 11/03/2010 00:58:25
Originally by: Dr BattleSmith

How long do you expect individuals to throw isk into product-lines which are flooded before realising the tool is burning their wallets?


I don't know, but i sure expect to find out. YARRRR!!

Edit: I actually expect the destructive influence on an item's profit margin to be ratehr long-lived, as it's more difficult and expensive to market-manipulate a price up, then it is to manipulate it down.

Dr BattleSmith
PAX Interstellar Services
Posted - 2010.03.11 01:14:00 - [123]
 

Originally by: Xenofur

Edit: I actually expect the destructive influence on an item's profit margin to be ratehr long-lived, as it's more difficult and expensive to market-manipulate a price up, then it is to manipulate it down.


It's also very easy to cancel the order, reprocess the flooded item, and make something else that will actually sell, even if isk is lost in the process.

The problem you have with your idea of "destruction of Jita" is that the tool would have to be sustainable to it's users.

With users quickly discovering that it leads them towards product-lines that are completely over-subscribed it's impact is reduced.

If it gave users profits on a sustainable basis then there may be some arguement for a larger impact on trading in general.
Without this sustainablilty there is no reason for users to continue to drive the market down as they can get better profits by *not* using the tool.

Xenofur
Aliastra
Posted - 2010.03.11 01:21:00 - [124]
 

Originally by: Dr BattleSmith

It's also very easy to cancel the order, reprocess the flooded item, and make something else that will actually sell, even if isk is lost in the process.


Interesting point. However that only applies if the person who invested the capital to drive the price down did so in a deliberate attempt to manipulate it and was acting alone. I found in my production career that the one thing driving prices down is idiots dumping massive amounts of stuff at ridiculously low prices and flat out leaving it there. Get hundreds of idiots doing the same thing and you have a deadlock.
Originally by: Dr BattleSmith

The problem you have with your idea of "destruction of Jita" is that the tool would have to be sustainable to it's users.

With users quickly discovering that it leads them towards product-lines that are completely over-subscribed it's impact is reduced.

If it gave users profits on a sustainable basis then there may be some arguement for a larger impact on trading in general.
Without this sustainablilty there is no reason for users to continue to drive the market down as they can get better profits by *not* using the tool.

Please, you've repeated the same thing 4 times now. I get what you're trying to say, and i concede it is a possibility, but neither of us can claim to know which actually *will* happen.

Dr BattleSmith
PAX Interstellar Services
Posted - 2010.03.11 01:26:00 - [125]
 

Originally by: Xenofur
Get hundreds of idiots doing the same thing and you have a deadlock.


This.

Xenofur
Aliastra
Posted - 2010.03.11 01:40:00 - [126]
 

Originally by: Dr BattleSmith
Originally by: Xenofur
Get hundreds of idiots doing the same thing and you have a deadlock.
This.
Just checking: You're agreeing with me that this will lock the price at a low level for long foreseeable times?

On a sidenote there: I now find myself thinking about what happens when traders pick up on the fact that there are MASSES of cheap items in Jita that will still pull good profits in other regions.

Dr BattleSmith
PAX Interstellar Services
Posted - 2010.03.11 02:06:00 - [127]
 

Edited by: Dr BattleSmith on 11/03/2010 02:11:06
Originally by: Xenofur
Just checking: You're agreeing with me that this will lock the price at a low level for long foreseeable times?


No I'm saying users of the tool will find themselves in these low profit items until they reprocess or stop following the crowd.

The low profits don't effect anyone except users of the tool who have bought into flooded markets.

For everyone else life goes on as usual while you play silly-buggers under-cutting each other over a few items no one cares about.

The only detrimental effect the tool has on the market is for those people using it to buy into that deadlock.

For the tool to have a wider impact on the market as a whole it would need to be "good" for it's users rather then bad.

Xenofur
Aliastra
Posted - 2010.03.14 13:09:00 - [128]
 

Edited by: Xenofur on 14/03/2010 13:15:17
Did a bunch of updates. Mostly cleaning up in the core and such, but I've also gotten around to setting up modifiable settings, so you can now have it calculate profits for your specific skill levels and most importantly:

You can now see profits for various BPO ME levels!

Edit: Profit calculation is a bit more accurate now too. Before this i was applying Accounting skill and broker fee to all incoming price values, which is wrong, since obviously materials bought to build stuff are bought for the plain price without any taxes.

Kazuo Ishiguro
House of Marbles
Posted - 2010.03.14 20:43:00 - [129]
 

It seems to use rather a lot of memory when sorting the table - firefox spiked to over 700mb. Could this be reduced for people using slower computers?

Xenofur
Aliastra
Posted - 2010.03.14 21:20:00 - [130]
 

Well, due to how HTML and Javascript works there aren't many ways to directly optimize this. In order to trade execution speed the sorting is done on a copy of the table, held in memory.

The only way to actually get it to use less memory is to reduce the amount of data.

Luckily I am planning to do that anyhow, by adding options to cut off items with a daily profit number below a certain amount of ISK, as well as those with a ROI above a certain number of days. I already have some in mind, but feel free to make suggestions as to the default values of those as well as other criterions that could be used to cut out irrelevant items. :)

Xenofur
Aliastra
Posted - 2010.03.14 22:39:00 - [131]
 

And preliminary filtering is in place. The price data for these items is still calculated and all, so by adjusting the filters they'll show up again and there's no way to miss anything, since the filtering is done after the calculations. Plus sides:

A good bunch less processing to do on a page request (the template was actually a huge part of that). Less bandwidth used (it used to be ~1mb per page. Sorting is now nearly instantaneous.

Mara Rinn
Posted - 2010.03.15 12:15:00 - [132]
 

Edited by: Mara Rinn on 15/03/2010 12:15:56
Change the profit threshold to a percentage, and you're golden :)

Does anyone care about the difference of 1000 ISK between two abaddons?

Does anyone not care about the difference of 10 ISK between two "Republic Fleet EMP S"?

Nice to see that Large Sentry Damage rigs are falling in price :)

Spruillo
Gallente
Spruillo Corp
Posted - 2010.03.15 14:05:00 - [133]
 

open markets grow closed markets suffocate jita will march on

Xenofur
Aliastra
Posted - 2010.03.15 14:55:00 - [134]
 

Edited by: Xenofur on 15/03/2010 15:02:29
Originally by: Mara Rinn

Change the profit threshold to a percentage, and you're golden :)

Nice idea, I wonder why i never thought of that. I added a filter for that and also display for how many items are hidden.

Edit: I wonder how interested people would be in getting the data also in JSON. (XML is for scrubs.)

Xenofur
Aliastra
Posted - 2010.03.19 20:56:00 - [135]
 

Just switched from the Eve-Metrics XML API to the JSON one. (Internal reasons mostly, faster to parse, less bandwidth used.) Please poke if anything behaves oddly.

Xenofur
Aliastra
Posted - 2010.03.20 13:23:00 - [136]
 

And some more options added:

You can now select the number of production slots you have available, so you'll be able to see exactly how much better running that battleship on ten slots at the same time is, when compared to the rig BPO in only one slot.

Amusingly the Megathron with 175M ISK profit per day, using 11 slots and a fully researched BPO, is still beat out by the Medium Core Defence Field Purger I with 216M ISK.

Due to a request i also added an option to display the production excess column again, though it defaults to off.

And lastly, since the coloring of sales vs. production was a bit skewed i completely reworked the calculations there so they now balance properly.

Xenofur
Aliastra
Posted - 2010.03.30 02:08:00 - [137]
 

Edited by: Xenofur on 30/03/2010 02:09:06
Reworked a whole bunch of the guts underneath in order to be able to multiplex the data sources easier. The final goal will be to be able to increase the day range for the history data to 90 days, but that'll have to wait a bit.

Immediate benefits for now will be that it tries to get as much sell and buy volume data from even central during normal operation and also retrieve old values if price updates from eve metrics come in. This means that the buy and sell vol numbers will never be particularly accurate, but at least they will be present most of the time so you have some idea about them.

If anything behaves oddly, please poke me, as it's 4 am now and i'm not sure how stable my changes were.

Lashmar
Posted - 2010.03.30 13:46:00 - [138]
 

Hello Xenofur

Thank you for the tool. I agree with a lot of the posts here, looking at the top of the list is not the way to go, but there is a lot of good information on your page none the less and its helped me show some big profits.

One thing i have often wondered is would it be possible to have a search function? It would be handy to know how well certain items you have the BPO researched for are doing if you have a choice, because looking through the list each time can get a bit much.

Thanks again for the tool, and consider this a free bumpVery Happy

Mingus Sin
Posted - 2010.03.30 15:55:00 - [139]
 

I'm curious as to how the manufacturing cost was reached. I see a few items that seem a bit on the low side. Are you using a perfect ME BPO for the calculations?

Xenofur
Aliastra
Posted - 2010.03.30 16:39:00 - [140]
 

Originally by: Lashmar
Hello XenofurThank you for the tool. I agree with a lot of the posts here, looking at the top of the list is not the way to go, but there is a lot of good information on your page none the less and its helped me show some big profits.
Thanks for the kind words. Smile If you have suggestions as how to improve the display to highlight "better" items, i'll be happy to hear them.
Originally by: Lashmar
One thing i have often wondered is would it be possible to have a search function? It would be handy to know how well certain items you have the BPO researched for are doing if you have a choice, because looking through the list each time can get a bit much.
I think right now the best way to find specific items is to ctrl+f for them.

Code-wise i have two things planned:

Highlighting BPOs. I already have some code in there that colours specific BPOs depending on whether they're unresearched, in progress or researched and which also alters the calculation to adapt the profits to the ME and PE of those BPOs. However that is very much a hack at the moment and will take some time until it can be considered production-ready.

I'm also planning to add some way to make it possible to select only specific blueprints to display.

If you have suggestions for either of these paths, feel free to voice them. :)


Originally by: Mingus Sin
I'm curious as to how the manufacturing cost was reached. I see a few items that seem a bit on the low side. Are you using a perfect ME BPO for the calculations?
Can you point out the specific items?

Manufacturing costs are reached by resolving the BPO build requirements and using the sell order value of the lowest 5% of the market as the price of the raw materials.

BPO ME can be customized in the customization block.

Mingus Sin
Posted - 2010.03.30 17:38:00 - [141]
 

Originally by: Xenofur


Manufacturing costs are reached by resolving the BPO build requirements and using the sell order value of the lowest 5% of the market as the price of the raw materials.

BPO ME can be customized in the customization block.


Let me double check before I give you any figures, I am pulling numbers from memory but with the way mineral prices have been falling recently, the figures I am thinking of could be way off. Thanks for the reply, the lowest 5% of the sell orders is exactly what I was looking for.

Awesome work.

Xenofur
Aliastra
Posted - 2010.03.30 19:01:00 - [142]
 

Thanks for the kind words and also the care taken. I can never guarantee 100% correctness, so it's good to have others double-check the data. :)

Xenofur
Aliastra
Posted - 2010.04.01 02:37:00 - [143]
 

Updated the market movement calculation today. Prices and market movements are now fetched separately.

Movements are fetched once at midnight and present an average of the past 90 days, which should make the value for possible sales per day MUCH more reliable and at the same time cuts down on the eve-metrics bandwidth i burn through.

If anyone has thoughts on that, don't hold back, same for when you notice any "off" values.

Xenofur
Aliastra
Posted - 2010.04.11 15:48:00 - [144]
 

I've now made the manufacture price a link that opens up a bill of materials, along with material prices, requirements and availability.

Incipus
Posted - 2010.04.12 15:20:00 - [145]
 

I remembered this appearing deceptively useful, but in practice worthless. I tested it on a random item.

Multifrequency L. Not the most profitable, not the least. A quick build cost calculation using my tool is 153k, you say 163k....A reasonable deviation.

A few points:
1. (OK)build cost - 153k(me) 163k(you)..Seems accurate.
2. (OK)sell revenue - 234k AVG Yesterday. You say 234k. OK, accurate.
3. (FAIL)Per unit profit 234-153 = 81k. How is there not a field for this? Seriously.
4. (FAIL)Daily sales volume 1134(you) last 7 days avg is 2904...Wrong.
5. (FAIL)Daily profit is 25.487M. Unit profit is 115k, that would involve selling 214 units, yet you say 360/day are produced and 1100/day are sold in jita. Huh? Where is this worthless figure from?

My advice:
1. Give a per hour profit.
2. Give a per unit profit.(How can you seriously not have this?)
3. Kill off the blueprint ROI crap.

Things you did well:
1. Build cost isn't unreasonable.
2. Using average sales value as the revenue.


Dav Varan
Posted - 2010.04.12 17:13:00 - [146]
 

Nice app.

Are you planning to extend it for T2 / T3 at any point ?

Xenofur
Aliastra
Posted - 2010.04.12 19:50:00 - [147]
 

Originally by: Incipus
I remembered this appearing deceptively useful, but in practice worthless.
Nah, i just thought a step further than you think i did. Wink

Originally by: Incipus
3. (FAIL)Per unit profit 234-153 = 81k. How is there not a field for this? Seriously.
Because i see no practical use for this to be displayed. Possible per-day profit and possible margin (far-left column) tell me all i need to know. Please tell me what you would need this value for. Smile

Originally by: Incipus
4. (FAIL)Daily sales volume 1134(you) last 7 days avg is 2904...Wrong.
I take the average of the last 90 days. Additionally, since i compare it to daily production, i need to adapt it a bit to more "real" sales numbers. (You'll NEVER sell a whole day's worth of items for an entire region, since you'd need to cover every station and have zero competitors, or play 0.01 isk all day.) As such, i simply slashed it in half. Thus you get that number there. It does not show how much is sold in the entire region, but how much you can reasonably expect to sell yourself.

Originally by: Incipus
5. (FAIL)Daily profit is 25.487M. Unit profit is 115k, that would involve selling 214 units, yet you say 360/day are produced and 1100/day are sold in jita. Huh? Where is this worthless figure from?
Please review your numbers. Single unit profit is calculated like this internally:
unit_profit = sell_price - manuf_cost - ( sell_price * broker_fee ) - ( sell_price * sales_tax )
70.7215k = 234k - 163k - ( 234k * 0.00039 ) - ( 234k * 0.0008 )

Day profit is unit_profit * min( daily_sales, daily_prod ).
25.46M = 70.7215k * min( 1134, 360 )

And there you go. It makes perfect sense.

Originally by: Incipus
1. Give a per hour profit.
Due to the fact that you can build 24 hours of the day, but only sell on 23 (well, more like 22) hours of the day, giving a per hour profit is kinda pointless. It is much more accurate to go by the day.

Originally by: Incipus
3. Kill off the blueprint ROI crap.
Huh? Why? It's useful to me. If you don't like it you can just set the minimum ROI to 99999 or whatever.

Originally by: Incipus
2. Using average sales value as the revenue.
I use something a lot better, as the EM guys are using a formula i supplied to them.

Anyhow, thanks for not telling me that everything sucked :D

Xenofur
Aliastra
Posted - 2010.04.12 19:51:00 - [148]
 

Originally by: Dav Varan
Nice app.

Are you planning to extend it for T2 / T3 at any point ?


I'm actually just filtering those right now as i don't plan to get into invention and all that, but i can see if i can cook up some nice way to make that filtering an option. :)


Pages: 1 2 3 4 [5]

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only