While I wish VS could be optimized further, its a sad pipe dream.
I don't think many understand the scope of a project like that. It would require a huge re-write of their backend.
VS is basically hitting a database with billions of records...the majority which (I would guess) are text strings... which from a DBA perspective, is extremely slow and taxing on server resources.
They would need to go in and index and/or add ID references for many items...probably 10,000+
and in some cases you couldn't even do that..because some text strings are user input (i.e. item descriptions,engravements, etc... I can't even think of all the scenarios.
I doubt it'll happen with their limited resources.
I don't think many understand the scope of a project like that. It would require a huge re-write of their backend.
VS is basically hitting a database with billions of records...the majority which (I would guess) are text strings... which from a DBA perspective, is extremely slow and taxing on server resources.
They would need to go in and index and/or add ID references for many items...probably 10,000+
and in some cases you couldn't even do that..because some text strings are user input (i.e. item descriptions,engravements, etc... I can't even think of all the scenarios.
Yeah, I don't think many realize just how big this would be or how complicated it would be. And we haven't even touched on all of the custom stuff like EM items (which are different between shards).
VS was a pleasant surprise when I came back, much better than the old third-party web-based search somebody came up with that used bots to scan select vendor NPCs (although that was nice to be able to look at stuff when you weren't playing and plan out things).
The game is a husk with a virtually non-existent development team. Cheaters run absolutely rampant, significant features lay around broken for years with no hope of repair, and the only significant client development of the last 10+ years has come at the hands of unpaid enthusiasts. Vendor search will never be fixed, it will sit there like a broken gangrenous limb forever. At least on the retail shards.
I've asked for vendor search to be fixed for years and have given up. They'd rather innovate ToTs to make the botters profit harder off them while discouraging legitimate players participation. Sad.
This discussion has been closed.
I will be slow to reply because I cannot log in/stay logged in to the forums. Make this your signature if you are tired of Vendor Search being broken, over FIVE YEARS and counting. Vendor search rendered useless after Publish 106 – Forsaken Foes on August 14, 2019.
Video games are meant to entertain, challenge, and immerse players in fascinating worlds. Unfortunately, Ultima Online fails to deliver on all fronts, making it a frustrating and forgettable experience.
One of the biggest issues with the game is its lackluster gameplay mechanics. Movement feels sluggish, combat is uninspired, and the progression system lacks meaningful rewards. Players often find themselves performing repetitive tasks with little excitement, turning what should be an engaging experience into a tedious grind.
The game’s graphics are equally disappointing. While modern titles boast breathtaking visuals and immersive environments, Ultima Online suffers from poorly rendered textures, awkward animations, and lifeless character models. Instead of creating an engaging world, the game presents an uninspired landscape that fails to captivate.
Adding to the frustration is the poorly written storyline. The game tries to weave a grand narrative but instead delivers a predictable and shallow plot with bland dialogue and forgettable characters. Players are left feeling detached from the story, making it difficult to care about the game's outcome.
Lastly, technical issues plague the experience. Frequent bugs, crashes, and glitches create an unstable gaming environment. Some features don't work as intended, leading to player dissatisfaction and disappointment.
In conclusion, Ultima Online is an unfortunate example of a game that fails in nearly every aspect. With dull gameplay, unimpressive visuals, a weak storyline, and numerous technical problems, it is hard to recommend to any gamer.
Just trying to fit in with the vibe y'all.
Just imagine if I went through life like this, lol.
The irony is exquisite.. he did his brick wall text in 15 seconds using AI.
Even when trolling.. he is cheating.
Just imagine if he went through life without cheating and lazy trolling like that, lol. PS: you wouldn't check any IRL Vibe test, with your mind set.. ever.. so don't bother trying to fit in.
So rather than recognise the effort the botters went to, to set all that up - for the benefit of the players, to help get certain items, something you could never be bothered to do, you would rather drag people backwards to your neanderthal world?
Leave attended paying accounts alone, these people go thru a lot of trouble to play/automate the game.
-UO official forums, brought to you by BoardSword studio
The irony is exquisite.. he did his brick wall text in 15 seconds using AI.
Even when trolling.. he is cheating.
Just imagine if he went through life without cheating and lazy trolling like that, lol. PS: you wouldn't check any IRL Vibe test, with your mind set.. ever.. so don't bother trying to fit in.
Haha, well spotted.
Just imagine if he went through life without cheating and lazy trolling like that, lol. Erm........
While I wish VS could be optimized further, its a sad pipe dream.
I don't think many understand the scope of a project like that. It would require a huge re-write of their backend.
VS is basically hitting a database with billions of records...the majority which (I would guess) are text strings... which from a DBA perspective, is extremely slow and taxing on server resources.
They would need to go in and index and/or add ID references for many items...probably 10,000+
and in some cases you couldn't even do that..because some text strings are user input (i.e. item descriptions,engravements, etc... I can't even think of all the scenarios.
I doubt it'll happen with their limited resources.
First off - You are likely correct in that they are loading and searching a bunch of text strings. That is super expensive on a database and searching.
However - its not billions of things. Do you know how much a billion is? We have about 10,000 players at the VERY BEST. That would require every single player to have vendors with 100,000 items PER PLAYER. Sorry - its not a billion or anywhere close to it. Probably more like a 2-3 million at best across all shards and as we know vendor search is shard specific.
What they should be doing is separating out item properties from text searches. They do *some* of that with being able to search by say 'chest piece' or 'helm' or whatever - but when they do this - it really messes up the search. As others have reported they have something going on with their SQL statement building that causes things to be missed when they are clearly in the database.
From a performance perspective you want indexed tables and you want ways to filter the records first.
Honestly there are a few main categories that would really cut down the database and be forced on a search : Armor, Jewelry, Weapon, PowerScroll, Alacrity, Transcendance for example. From there everything else is in an 'other' category. That would start separating out the 'buckets' of stuff and honestly those could all be in separate tables. Then when you search you are FORCED to select one of these main categories first then you are property/text searching across way less items. I never really want a result set that includes weapons, rings, braces, alacrity, PS etc etc when I'm after say a 110 Necromancy Scroll.
I know you CAN subselect some of these now - but you aren't forced to which makes lazy people just type in 'Necromancy' and get everything under the sun with Necro on it. Rings, Books, scrolls etc. These become expensive and consuming searches on the DB. They don't need to be.
Separate into several tables. Put a drop down to start the search and pick a main category. From there then you can property and text string search. This is how it should work.
And into todays modern world of high performance databases and big data/no sql/ hadoop and the like - you can handle massive amounts of information cheaply and easily. I got time - i'll rewrite vendor search. I only spent 30 years in data.
Comments
VS was a pleasant surprise when I came back, much better than the old third-party web-based search somebody came up with that used bots to scan select vendor NPCs (although that was nice to be able to look at stuff when you weren't playing and plan out things).
I will be slow to reply because I cannot log in/stay logged in to the forums.
Make this your signature if you are tired of Vendor Search being broken, over FIVE YEARS and counting.
Vendor search rendered useless after Publish 106 – Forsaken Foes on August 14, 2019.
Video games are meant to entertain, challenge, and immerse players in fascinating worlds. Unfortunately, Ultima Online fails to deliver on all fronts, making it a frustrating and forgettable experience.
One of the biggest issues with the game is its lackluster gameplay mechanics. Movement feels sluggish, combat is uninspired, and the progression system lacks meaningful rewards. Players often find themselves performing repetitive tasks with little excitement, turning what should be an engaging experience into a tedious grind.
The game’s graphics are equally disappointing. While modern titles boast breathtaking visuals and immersive environments, Ultima Online suffers from poorly rendered textures, awkward animations, and lifeless character models. Instead of creating an engaging world, the game presents an uninspired landscape that fails to captivate.
Adding to the frustration is the poorly written storyline. The game tries to weave a grand narrative but instead delivers a predictable and shallow plot with bland dialogue and forgettable characters. Players are left feeling detached from the story, making it difficult to care about the game's outcome.
Lastly, technical issues plague the experience. Frequent bugs, crashes, and glitches create an unstable gaming environment. Some features don't work as intended, leading to player dissatisfaction and disappointment.
In conclusion, Ultima Online is an unfortunate example of a game that fails in nearly every aspect. With dull gameplay, unimpressive visuals, a weak storyline, and numerous technical problems, it is hard to recommend to any gamer.
Just trying to fit in with the vibe y'all.
Just imagine if I went through life like this, lol.
Even when trolling.. he is cheating.
Just imagine if he went through life without cheating and lazy trolling like that, lol.
PS: you wouldn't check any IRL Vibe test, with your mind set.. ever.. so don't bother trying to fit in.
Leave attended paying accounts alone, these people go thru a lot of trouble to play/automate the game.
-UO official forums, brought to you by BoardSword studio
Just imagine if he went through life without cheating and lazy trolling like that, lol.
Erm........
You are right, I never bothered trying to fit in.
However - its not billions of things. Do you know how much a billion is? We have about 10,000 players at the VERY BEST. That would require every single player to have vendors with 100,000 items PER PLAYER. Sorry - its not a billion or anywhere close to it. Probably more like a 2-3 million at best across all shards and as we know vendor search is shard specific.
What they should be doing is separating out item properties from text searches. They do *some* of that with being able to search by say 'chest piece' or 'helm' or whatever - but when they do this - it really messes up the search. As others have reported they have something going on with their SQL statement building that causes things to be missed when they are clearly in the database.
From a performance perspective you want indexed tables and you want ways to filter the records first.
Honestly there are a few main categories that would really cut down the database and be forced on a search : Armor, Jewelry, Weapon, PowerScroll, Alacrity, Transcendance for example. From there everything else is in an 'other' category. That would start separating out the 'buckets' of stuff and honestly those could all be in separate tables. Then when you search you are FORCED to select one of these main categories first then you are property/text searching across way less items. I never really want a result set that includes weapons, rings, braces, alacrity, PS etc etc when I'm after say a 110 Necromancy Scroll.
I know you CAN subselect some of these now - but you aren't forced to which makes lazy people just type in 'Necromancy' and get everything under the sun with Necro on it. Rings, Books, scrolls etc. These become expensive and consuming searches on the DB. They don't need to be.
Separate into several tables. Put a drop down to start the search and pick a main category. From there then you can property and text string search. This is how it should work.
And into todays modern world of high performance databases and big data/no sql/ hadoop and the like - you can handle massive amounts of information cheaply and easily. I got time - i'll rewrite vendor search. I only spent 30 years in data.