DC
Not too sure exactly when but I think it was sometime in 2009.
Cool, that's the one I'm working on now. Been looking for some reference for that studio so thanks for posting that image. It'll come in very handy as a lot of the youtube videos featuring this studio don't really show the wood grain very clearly
My favourite studio is the one with the warm wood panels; arguably have the most memories from that period with the hilarity of the introduction of speed auctions.
Cool, that's the one I'm working on now. Been looking for some reference for that studio so thanks for posting that image. It'll come in very handy as a lot of the youtube videos featuring this studio don't really show the wood grain very clearly
Last edited by DanielC on 23 November 2016 12:41am
TA
As a tech lead C# developer myself (day job), and having some experience in Unity; have you never considered importing your studio models into Unity; and having a full space 3D studio; with set camera tracks, zooms, pans, etc? It shouldn't be too much of a stretch from what you've already accomplished; given the amount of effort you've put into designing the studios.
'Course, I'm assuming you've modelled them in a format that's compatible with Unity.
It'd be really cool to pan the camera from a pod to the gavel for example with Rising Price auctions; and perform slow zooms towards the closing of auctions.
Also, a question on the simulation logic; do user-actions have an impact on how frequently new bids are placed; for example - does showing the "outbid" panel cause a slight weight increase in the probability of a new bid being placed? Does showing "top bidders" cause a slight weight increase in the probability of new bids being higher? Does switching camera from gavel to center to product back to center increase the probability of a surge of last minute bids?; and same question for a clock as the clock runs down - is the weighted probability of new bids increased; specially if the scale of the saving against the guide price is large and a clock is about to expire?
I've also noticed a couple of primitive logic flaws. The frequency of bids for an item decreases dramatically if there is a lower quantity ( I'm not sure quantity should be used in your weighted logic -- assuming you're using shifting weights to determine the frequency and cap of bids).
Items with lesser quantities also seem to reach their target close price consistently - but higher quantities rarely do; and this is a problem with using quantity in determining value (and frequency) of bids.
If an auction is due to close (timed perhaps) and is drastically below the "unit cost"; and even further below the guide price by a significant amount - a surge of last 60 second bidders would offer far more realism.
From my observations, at the moment, the probability of bids being placed at all seems based on the 'saving' and the 'quantity'; time is not a factor - and nor is user input in controlling the auction. The quantity factor is the biggest hindrance in reaching something closer to the real thing. Happy to be corrected though!
I work with these types of numbers on a daily basis in my day job - so thought I'd offer some input and ask some questions
'Course, I'm assuming you've modelled them in a format that's compatible with Unity.
It'd be really cool to pan the camera from a pod to the gavel for example with Rising Price auctions; and perform slow zooms towards the closing of auctions.
Also, a question on the simulation logic; do user-actions have an impact on how frequently new bids are placed; for example - does showing the "outbid" panel cause a slight weight increase in the probability of a new bid being placed? Does showing "top bidders" cause a slight weight increase in the probability of new bids being higher? Does switching camera from gavel to center to product back to center increase the probability of a surge of last minute bids?; and same question for a clock as the clock runs down - is the weighted probability of new bids increased; specially if the scale of the saving against the guide price is large and a clock is about to expire?
I've also noticed a couple of primitive logic flaws. The frequency of bids for an item decreases dramatically if there is a lower quantity ( I'm not sure quantity should be used in your weighted logic -- assuming you're using shifting weights to determine the frequency and cap of bids).
Items with lesser quantities also seem to reach their target close price consistently - but higher quantities rarely do; and this is a problem with using quantity in determining value (and frequency) of bids.
If an auction is due to close (timed perhaps) and is drastically below the "unit cost"; and even further below the guide price by a significant amount - a surge of last 60 second bidders would offer far more realism.
From my observations, at the moment, the probability of bids being placed at all seems based on the 'saving' and the 'quantity'; time is not a factor - and nor is user input in controlling the auction. The quantity factor is the biggest hindrance in reaching something closer to the real thing. Happy to be corrected though!
I work with these types of numbers on a daily basis in my day job - so thought I'd offer some input and ask some questions
Last edited by Tao on 23 November 2016 11:52am - 4 times in total
DC
Sure no problem. I've uploaded a little video showing some of the features of this graphics set, and the new timer on the earlier graphics.
It's something I considered right back at the start of the project, but in the end decided to settle on just using pre-rendered images. It's something I could maybe experiment with for future versions, however the models will need re-worked to be more game-engine friendly since at the moment they're all high-poly subdivision surface.
The user actions don't have any impact on bids being placed, but some good ideas here I can definitely try to add for next update.
Yeah lower quantities in the sim mean less chance of bids being placed. The reason I did that was because I wasn't sure of the best way to stop lower quantities from reaching overly high prices, also for falling price auctions I didn't want quantities of 1000 taking far too long to sell, since viewers also lose interest as the time elapsed increases.
I'll need to have another think about how to make it behave more realistically. I've never done any programming before this project so I may have a few questions for you when I get on to it
Any chance of seeing how the chevron graphics will work? I'm guessing we'll able to put our own custom text in after the 'Live Event' text appears?
Sure no problem. I've uploaded a little video showing some of the features of this graphics set, and the new timer on the earlier graphics.
As a tech lead C# developer myself (day job), and having some experience in Unity; have you never considered importing your studio models into Unity; and having a full space 3D studio; with set camera tracks, zooms, pans, etc? It shouldn't be too much of a stretch from what you've already accomplished; given the amount of effort you've put into designing the studios.
'Course, I'm assuming you've modelled them in a format that's compatible with Unity.
It'd be really cool to pan the camera from a pod to the gavel for example with Rising Price auctions; and perform slow zooms towards the closing of auctions.
'Course, I'm assuming you've modelled them in a format that's compatible with Unity.
It'd be really cool to pan the camera from a pod to the gavel for example with Rising Price auctions; and perform slow zooms towards the closing of auctions.
It's something I considered right back at the start of the project, but in the end decided to settle on just using pre-rendered images. It's something I could maybe experiment with for future versions, however the models will need re-worked to be more game-engine friendly since at the moment they're all high-poly subdivision surface.
Also, a question on the simulation logic; do user-actions have an impact on how frequently new bids are placed; for example - does showing the "outbid" panel cause a slight weight increase in the probability of a new bid being placed? Does showing "top bidders" cause a slight weight increase in the probability of new bids being higher? Does switching camera from gavel to center to product back to center increase the probability of a surge of last minute bids?; and same question for a clock as the clock runs down - is the weighted probability of new bids increased; specially if the scale of the saving against the guide price is large and a clock is about to expire?
The user actions don't have any impact on bids being placed, but some good ideas here I can definitely try to add for next update.
I've also noticed a couple of primitive logic flaws. The frequency of bids for an item decreases dramatically if there is a lower quantity (
I'm not sure quantity should be used in your weighted logic
-- assuming you're using shifting weights to determine the frequency and cap of bids).
Items with lesser quantities also seem to reach their target close price consistently - but higher quantities rarely do; and this is a problem with using quantity in determining value (and frequency) of bids.
If an auction is due to close (timed perhaps) and is drastically below the "unit cost"; and even further below the guide price by a significant amount - a surge of last 60 second bidders would offer far more realism.
From my observations, at the moment, the probability of bids being placed at all seems based on the 'saving' and the 'quantity'; time is not a factor - and nor is user input in controlling the auction. The quantity factor is the biggest hindrance in reaching something closer to the real thing. Happy to be corrected though!
I work with these types of numbers on a daily basis in my day job - so thought I'd offer some input and ask some questions
Items with lesser quantities also seem to reach their target close price consistently - but higher quantities rarely do; and this is a problem with using quantity in determining value (and frequency) of bids.
If an auction is due to close (timed perhaps) and is drastically below the "unit cost"; and even further below the guide price by a significant amount - a surge of last 60 second bidders would offer far more realism.
From my observations, at the moment, the probability of bids being placed at all seems based on the 'saving' and the 'quantity'; time is not a factor - and nor is user input in controlling the auction. The quantity factor is the biggest hindrance in reaching something closer to the real thing. Happy to be corrected though!
I work with these types of numbers on a daily basis in my day job - so thought I'd offer some input and ask some questions
Yeah lower quantities in the sim mean less chance of bids being placed. The reason I did that was because I wasn't sure of the best way to stop lower quantities from reaching overly high prices, also for falling price auctions I didn't want quantities of 1000 taking far too long to sell, since viewers also lose interest as the time elapsed increases.
I'll need to have another think about how to make it behave more realistically. I've never done any programming before this project so I may have a few questions for you when I get on to it
TA
I'm happy to talk sim logic with you in great technical detail/examples to help you improve this app; it's great fun and I think you've done a phenomenal job so far. The AI logic could be greatly improved though.
The only part I disagree with is that viewers lose interest over duration - shopping channels exist for channel hoppers; chances are you lose the vast majority of viewers you had at the start of the auction; and gain a completely new set of the same quantity by the middle or end of the auction; of course the exception are the few die hard fans; and those involved in the current sale.
Keep up the amazing work! Can't wait to see the new studios ^_^
Yeah lower quantities in the sim mean less chance of bids being placed. The reason I did that was because I wasn't sure of the best way to stop lower quantities from reaching overly high prices, also for falling price auctions I didn't want quantities of 1000 taking far too long to sell, since viewers also lose interest as the time elapsed increases.
I'm happy to talk sim logic with you in great technical detail/examples to help you improve this app; it's great fun and I think you've done a phenomenal job so far. The AI logic could be greatly improved though.
The only part I disagree with is that viewers lose interest over duration - shopping channels exist for channel hoppers; chances are you lose the vast majority of viewers you had at the start of the auction; and gain a completely new set of the same quantity by the middle or end of the auction; of course the exception are the few die hard fans; and those involved in the current sale.
Keep up the amazing work! Can't wait to see the new studios ^_^
DC
Yeah you'll be able to spawn/de-spawn the timer before starting the countdown. You'll also be able to enter your own custom text into the animated DOG, so it doesn't have to say 'smashing prices' etc.
That would be a great help, and very much appreciated! I'm gonna try and get the new studios finished before tackling this, but that would be very helpful
Very accurate recreations of all graphics and sets here! Can't wait to use the chevron graphics, I liked that set. Will the Spawn Timer button be available in the finished update?
Yeah you'll be able to spawn/de-spawn the timer before starting the countdown. You'll also be able to enter your own custom text into the animated DOG, so it doesn't have to say 'smashing prices' etc.
I'm happy to talk sim logic with you in great technical detail/examples to help you improve this app; it's great fun and I think you've done a phenomenal job so far. The AI logic could be greatly improved though.
That would be a great help, and very much appreciated! I'm gonna try and get the new studios finished before tackling this, but that would be very helpful
JB
Yeah you'll be able to spawn/de-spawn the timer before starting the countdown. You'll also be able to enter your own custom text into the animated DOG, so it doesn't have to say 'smashing prices' etc.
Nice touch there with the DOG! It would be great to enter some custom text to make up our own promotions etc.
Do you have the two minute version of the clock music? There was a two minute version available to listen online.
Very accurate recreations of all graphics and sets here! Can't wait to use the chevron graphics, I liked that set. Will the Spawn Timer button be available in the finished update?
Yeah you'll be able to spawn/de-spawn the timer before starting the countdown. You'll also be able to enter your own custom text into the animated DOG, so it doesn't have to say 'smashing prices' etc.
Nice touch there with the DOG! It would be great to enter some custom text to make up our own promotions etc.
Do you have the two minute version of the clock music? There was a two minute version available to listen online.
CY
cyberdude
How about adding the 'megadrop' VO as a sound effect?
JB
It doesn't need to be added as a separate sound effect. They only ever used it when a Megadrop occurred and you can already use the effect more than once.
How about adding the 'megadrop' VO as a sound effect?
It doesn't need to be added as a separate sound effect. They only ever used it when a Megadrop occurred and you can already use the effect more than once.