You can find a general description of the various network challenges FD's devs are facing in this AWS conference. I'm pretty confident that with your expertise, you'll be able to help them with these 15 years old DB technologies they use.
Is there a way to know what I have discovered first?
- Thread starter Pomerlaw
- Start date
You can find a general description of the various network challenges FD's devs are facing in this AWS conference. I'm pretty confident that with your expertise, you'll be able to help them with these 15 years old DB technologies they use.
I have some of mine jotted down on paper; E: D is in fact the first game I've played with a notebook and pen at the side of me for probably 20 years. It doesn't change the fact that I would much rather have the information accessible from within the game.
- - - - - Additional Content Posted / Auto Merge - - - - -
I remember the first time I decided to take a hi def one. Checked out image properties and I was like 'Oh 75MB, yeah won't be doing so many of those then...'
+1
And the longer they wait, the worse it gets.
Although, to be fair, this is probably not the number one feature that the majority of the player base is waiting for.
Do you mean as and when a particular player looks at a system map? That would work, but still relies on players knowing which systems they tagged so they can open the local map and update the local database.
If you mean automatically, then it would have the same initial problem of accessing the FD databases. As I understand it the only time the "first discovery" data is looked up is when a player opens a system map. To build up a local database of discoveries would require a client to simulate the opening of 400 billion system maps to see whether the current player had any tags in them. At one map per second that would take 12 years. Per player.
Unless there is a reverse lookup feature in the database(s) that wasn't explicit from the relevant part of the AWS video, there may not be a way around this.
(The key to this whole thing is the S3 database referenced here:
The easiest way to store a local database on the player's machine going forward would be to parse the "CONGRATULATIONS CMDR {name} YOU ARE FIRST TO DISCOVER..." pop-ups that appear when you sell the data, which is something I and no doubt many others currently do manually with good old pen and paper, or a spreadsheet. But that would still have the problem of only working for systems discovered after the feature was implemented.
I hope I'm wrong on all of this and there's an easy way for FD to implement an historical lookup of tagged systems, but I can't help but feel that if it was possible to do this it would have been done already. Then again that could be said of a great many things in this game that did eventually materialise, so maybe there's hope.
Last edited:
My method of keeping track is probably a bit time intensive. When exploring and finding virgin systems or planets, I screenshot the Sys Map and place it in a folder called 'Tentative' since its not mine til I sell the data and get the first discoverer tag, and it may be days before I get back, and of course someone might get it before I do. But then, after I sell my data, I go thru each pic and take another screenshot of each object that has my name. Then I list each object and type of object in a word pad document, too include System name. Then I delete all the screenshots, at 6 mb a screenshot, I would soon run out of room on my hard drive unless I converted the shots to another format. I could literally list every object in ED in a word document and it might reach 6 mb. Like I said, its time consuming, been on 3 trips out and so far have 1897 objects, and around 160 Systems that are completely mine. My goal is to name 1 million objects someday, so I don't care if its an icy rock that no one wanted to travel 500k ls to name, if I jump into a system and it is unnamed, I will scan everything in the system, cept the asteroidbelts. I have only been able to name one earth like so far, and a few water worlds and pretty much everything else except a black hole. But those 3 trips have only been learning trips, only went out 1000 or so lys, planning on a big trip soon. Gotta find that one earth like with sentient life that Devs say is out there.
Well I would hope not too many people are writing it on the back of a shopping list with your eldest daughter's lipstick.
Evernote. I keep a list of earth likes and ammonia worlds I discovered, just in case they become interesting later.
I do wish there was a way to query the game servers. If FD provided that I'm sure plenty of people would write tools for it.
I do wish there was a way to query the game servers. If FD provided that I'm sure plenty of people would write tools for it.
![[VR] Mad Mike](https://forums-cdn.zaonce.net/data/avatars/m/27/27910.jpg?1553527108){kind=avatar}
I wish we could. I had every single discovered 1st logged in my notes on slopeys tool.... And forgot to back up when I upgraded. Gutted as I wanted to go back and land on them. Once 2.1 is out going forward at least it will be fixed
While there are definitely more pressing issues, back when they decided on the `First Discovered By` mechanic, they must have surely foreseen that players would want to be able to look back on those discoveries, retrace journeys and such. Perhaps they have designed for that and have just not yet implemented it, but I'll admit I'm doubtful. It would be a pretty poor oversight on their part if they did not see how we might like this information.
Anyway, whilst we are dreaming, I'd like to see filters on the galaxy map for:
- First discovered systems.
- Systems visited, so I can visualise my exploration trips.
Further filters would of course be nice, earth likes etc...
Some kind of exploration archive should have been in the game since day 1 or at the very least since they added the 'discovered by' tags. It's a pretty big oversight IMO and it's sad that we don't even know if such a feature will ever be implemented. The fact that only now we're getting simple bookmarks isn't a good sign...
Last edited:
Do you mean as and when a particular player looks at a system map? That would work, but still relies on players knowing which systems they tagged so they can open the local map and update the local database.
If you mean automatically, then it would have the same initial problem of accessing the FD databases. As I understand it the only time the "first discovery" data is looked up is when a player opens a system map. To build up a local database of discoveries would require a client to simulate the opening of 400 billion system maps to see whether the current player had any tags in them. At one map per second that would take 12 years. Per player.
Unless there is a reverse lookup feature in the database(s) that wasn't explicit from the relevant part of the AWS video, there may not be a way around this.
(The key to this whole thing is the S3 database referenced here:
If it's possible to search that on the "cmdrname" field then we're golden and it's just a question of waiting for FD to implement the code in the game. But if, as I interpreted the video, those values can only be obtained if you already know the system name, then we'd have to look up every system which is impractical.)
The easiest way to store a local database on the player's machine going forward would be to parse the "CONGRATULATIONS CMDR {name} YOU ARE FIRST TO DISCOVER..." pop-ups that appear when you sell the data, which is something I and no doubt many others currently do manually with good old pen and paper, or a spreadsheet. But that would still have the problem of only working for systems discovered after the feature was implemented.
I hope I'm wrong on all of this and there's an easy way for FD to implement an historical lookup of tagged systems, but I can't help but feel that if it was possible to do this it would have been done already. Then again that could be said of a great many things in this game that did eventually materialise, so maybe there's hope.
No, it wouldn't require opening up 400 billion system maps.
Make a quick batch file that goes through all of the systems tagged as being discovered or having a discovered object. Make individual lists for each unique commander name containing the objects they've discovered. When that player logs in, they download their individual list which is then stored locally and kept up to date by fact-checking the list say..... Every 30 minutes, which takes the same bandwidth and server runtime as a latency ping. People typically don't sell exploration data very often and they sell it in large chunks so long intervals would be dandy. All of the scan information for objects on that list are then saved to the local machine and when the player opens up their exploration journal it is appropriately presented and available with no server interaction.
All of the exploration data is saved in text format, as the picture you linked displays. Each entry is a small paragraph of basic facts and a couple PG codes. The game doesn't work through the UI to find and parse the exploration data, that would be absolute idiocy. It just references the files on the servers to fill in the blank spots in the map UI.
Stop thinking within the confines of the games as it exists. It's still in development. They can add simple server functions and in-game features as they please. Making a single button in the exploration journal UI would take more time than writing said batch file. You're adding fantastical complications where they don't exist.
Edit: The operation wouldn't cause some severe server overload either. They already run several similar operations in the background to mine the exploration database for player behavior. You'd just be adding one more function on top of the pile.
Last edited:
Again, I think that's where the problem lies. Right now there may not be a subset of systems "tagged as being discovered or having a discovered object" that can be "gone through." Just chunks of text, including discovery status, that are indexed solely by system.
You could go through that system by system but you'd have to do it for 400 billion records. How quickly you could do that only FD knows, and it certainly wouldn't be efficient. But maybe it would be worth doing as a one-off to reindex the data into a form more readily searchable by discovery status.
But only if the data already exist in a form that allows such functions to be run in reasonable time. It may already, or it may require the whole database to be read and restructured into a more conveniently searchable form for future use. We just don't know.
Look, I hope you're right and I'm wrong, I really do. But people are asking why this hasn't been implemented and I'm suggesting that it might be because of the short-sighted way in which the exploration database has been set up. Depending on interpretation, what's in that video may back that up or it may not. But it's the best resource we as users have. Only FD knows what's really going on. And despite their track record of heel-dragging on features I do worry that, in this particular case, if it was a SMOP we'd already have it in the game.
I'm not "adding fantastical complications." I'm stating things as I see them based on what little information we have. If you know otherwise please educate me. Otherwise you're speculating just as much as I am.
Chunks of text is all that a batch file needs. Humans aren't the only thing that can Ctrl+F.
My processor runs over 320 GFLOPS, which is enough to search the entire database in less than a second in a perfectly parallel operation. Frontier is renting dozens of servers which have at least 50% more GFLOPS available, assuming that they don't use GPU's, in which case it would be several times what my processor can handle. Per server. These functions can be run in the background and be completed in less than a day. After that it's background maintenance, which wouldn't increase any costs because you'd be cutting down on server load over the long term.
Go do a search on your computer for a file name.
Takes about 2 minutes? .001 seconds of that was processor time checking your indexed database, the rest of the time was spent waiting for your slow peasant hard drive to fact check it. Servers are orders of magnitudes more efficient on pulling stored information and checking databases. It's really a non-issue.
Sarcasm and misrepresentation of my statement noted.
The fact that the GAME accesses the same data certainly means that it is not an insurmountable task. If the insurance industry can do it with millions of cross linked records for things like life/auto/home insurance, certainly Frontier can do it for us. Whether it is something they are willing to invest time and resources in is a different issue.
From the perspective of the database server, however, it is just one more user asking for the same old data.
Me Too! Now do Video for the hand in's.
