• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

ARCHIVE - *** NVIDIA GEFORCE RTX 3080 SERIES STOCK SITUATION *** - NO COMPETITOR DISCUSSION

Status
Not open for further replies.
Kungfudave said:
]Im getting some pretty good clocks on my palit on std cooler. needs some tweaking though

Click to expand...
Well, synthetics can pass the tests without showing issues. Can you run some in-game benchmarks and see if it CTD or not?
I sincerely hope Palit didn't cheap out on the arrays.
 
NikTheSHNIK said:
As each order payment took a different time to process depending upon payment type/terms their order number doesn't match up to the times that orders were taken.

At that point we can no longer use a difference in order number to determine a place in queue.

Add into that the fact we were still processing payments and cancellations until yesterday (I was helping out but there was still webnotes to clear from days before) an individual order's queue position was changing hourly, therefore an accurate queue position was impossible to convey.

For example there were a ton of American customers who'd ordered not realizing they'd have to pay sales tax. Each one had to contact us and tell us they'd cancel and each one bumped someone else up a place in the queue.

We tried to get through all the queries as quickly as possible. Hopefully someone waiting a day/few days after their order and being able to give them an accurate idea of their order position (after we'd cleared those wanting to cancel and all those whose payments hadn't cleared), is better than giving them an inaccurate queue position.

Hopefully at that point we'd also have more info from Nvidia and the board partners about inventory/etas and be able to provide a more accurate queue position/stock eta.

Sometimes these things take days rather than hours.

I know one of the sales guys answered over 300 webnotes about 3080 orders one day, on top of the normal workload! That's 1.6 webnotes a minute that day, and we were all hands to the pump between us to help out and answer them. There was a couple of days I answered 100 3080 queries and I'm supposed to be managing peripherals and monitors!
Click to expand...

Finally some degree of clarification. Thank you Nik. Now why couldn't Gibbo have just said this a couple of days ago, there'd have been less drama?!?!? I will now sit and wait patiently, all I wanted was to understand WTF was going on.
 
Really pleased to see some proper OCUK interaction, even though its not from someone who is responsible for cards. This is what we want, communication, understanding, shame its taken so long but thanks!
 
NikTheSHNIK said:
As each order payment took a different time to process depending upon payment type/terms their order number doesn't match up to the times that orders were taken.

At that point we can no longer use a difference in order number to determine a place in queue.

Add into that the fact we were still processing payments and cancellations until yesterday (I was helping out but there was still webnotes to clear from days before) an individual order's queue position was changing hourly, therefore an accurate queue position was impossible to convey.

For example there were a ton of American customers who'd ordered not realizing they'd have to pay sales tax. Each one had to contact us and tell us they'd cancel and each one bumped someone else up a place in the queue.

We tried to get through all the queries as quickly as possible. Hopefully someone waiting a day/few days after their order and being able to give them an accurate idea of their order position (after we'd cleared those wanting to cancel and all those whose payments hadn't cleared), is better than giving them an inaccurate queue position.

Hopefully at that point we'd also have more info from Nvidia and the board partners about inventory/etas and be able to provide a more accurate queue position/stock eta.
Click to expand...
Thank you for taking time to write and tell us what is all about! It's really helpful and actually calming in a way.
 
ceekay said:
Well, synthetics can pass the tests without showing issues. Can you run some in-game benchmarks and see if it CTD or not?
I sincerely hope Palit didn't cheap out on the arrays.
Click to expand...
its stable on synthetic benches but u have to drop it down slightly for a 3d mark to around 2150. I haven't played with power delivery yet and i'm gonna wait to water block it as it started getting up to 75 degrees so I'm gonna wait till i have my block to push it. i dont like keeping the cards too hot. I think i can just about get it there stable. i got my palit 1080ti to run 2150mhz stable which was quite high.
From the images I've seen of teardowns they have got for a 5 and 1 setup on the power conditioning.
 
I was bored, and after @Soliddus mentioned that one of OC's support team said over the phone that they had sold about 1800 non-OC TUF cards, or about a third LESS than the spreadsheet (and I think unofficial queue calculator) are using to build their estimates, I though I might redraw some graphs based on the spreadsheet with the total sales changed to match that "known" data point. Then google broke the copy of spreadsheet I used to draw graphs last time (the tab just freezes if I try to load it), so I decided to redo the calculations from the fresh spreadsheet data, and do it properly.

So, new graphs:
  • They are working on the basis that in total OC sold about 10500 cards, not the 16000 the main spreadsheet uses
  • They calculate the number sold based on a progressively generated percentage of sales, rather than the overall percentage, which should make it more accurate
  • I've created graphs for all data, first 12 hours, first 3 hours, and first hour
    • Each of those then has a All Cards, Popular Cards (All 4 Asus, EVGA Black, Gigabyte Gaming and Eagle, Both MSI, and the Zotac Trinity) and Less Popular Cards (EVGA FTW Gaming & Ultra, EVGA XC3 Gaming & Ultra, Gigabyte AORUS Master & Xtreme, Inno3D Chill x2 & x3, KFA2 SG, Palit Gaming & OC
  • I've filtered the data so the cards with too early or too late (the future) aren't counted at all, since if people can't fill in a date correctly, I don't trust them to know what card they bought
  • They should automatically update the data based on new submission to the main spreadsheet
I found it interesting at least as a data point to compare the unofficial calculator to just to see how different processing of the data changes how far back you might be in the queue. If anyone else is interested they can find them here: https://docs.google.com/spreadsheets/d/1L5FsPk6ZsYzN8J2XMaAoLQPbCrjKBTiCEanh1AthKs0
 
Lastof said:
I was bored, and after @Soliddus mentioned that one of OC's support team said over the phone that they had sold about 1800 non-OC TUF cards, or about a third LESS than the spreadsheet (and I think unofficial queue calculator) are using to build their estimates, I though I might redraw some graphs based on the spreadsheet with the total sales changed to match that "known" data point. Then google broke the copy of spreadsheet I used to draw graphs last time (the tab just freezes if I try to load it), so I decided to redo the calculations from the fresh spreadsheet data, and do it properly.

So, new graphs:
  • They are working on the basis that in total OC sold about 10500 cards, not the 16000 the main spreadsheet uses
  • They calculate the number sold based on a progressively generated percentage of sales, rather than the overall percentage, which should make it more accurate
  • I've created graphs for all data, first 12 hours, first 3 hours, and first hour
    • Each of those then has a All Cards, Popular Cards (All 4 Asus, EVGA Black, Gigabyte Gaming and Eagle, Both MSI, and the Zotac Trinity) and Less Popular Cards (EVGA FTW Gaming & Ultra, EVGA XC3 Gaming & Ultra, Gigabyte AORUS Master & Xtreme, Inno3D Chill x2 & x3, KFA2 SG, Palit Gaming & OC
  • I've filtered the data so the cards with too early or too late (the future) aren't counted at all, since if people can't fill in a date correctly, I don't trust them to know what card they bought
  • They should automatically update the data based on new submission to the main spreadsheet
I found it interesting at least as a data point to compare the unofficial calculator to just to see how different processing of the data changes how far back you might be in the queue. If anyone else is interested they can find them here: https://docs.google.com/spreadsheets/d/1L5FsPk6ZsYzN8J2XMaAoLQPbCrjKBTiCEanh1AthKs0
Click to expand...

Awesome, just had a look and it's really interesting, thanks for taking the time to do this!

You can tell how many people there are lurking on the thread by looking at whos viewing the spreadsheet lol
 
I just checked MSI trio Gaming X, and they have unfortunately cheap MLCC
Lastof said:
I was bored, and after @Soliddus mentioned that one of OC's support team said over the phone that they had sold about 1800 non-OC TUF cards, or about a third LESS than the spreadsheet (and I think unofficial queue calculator) are using to build their estimates, I though I might redraw some graphs based on the spreadsheet with the total sales changed to match that "known" data point. Then google broke the copy of spreadsheet I used to draw graphs last time (the tab just freezes if I try to load it), so I decided to redo the calculations from the fresh spreadsheet data, and do it properly.

So, new graphs:
  • They are working on the basis that in total OC sold about 10500 cards, not the 16000 the main spreadsheet uses
  • They calculate the number sold based on a progressively generated percentage of sales, rather than the overall percentage, which should make it more accurate
  • I've created graphs for all data, first 12 hours, first 3 hours, and first hour
    • Each of those then has a All Cards, Popular Cards (All 4 Asus, EVGA Black, Gigabyte Gaming and Eagle, Both MSI, and the Zotac Trinity) and Less Popular Cards (EVGA FTW Gaming & Ultra, EVGA XC3 Gaming & Ultra, Gigabyte AORUS Master & Xtreme, Inno3D Chill x2 & x3, KFA2 SG, Palit Gaming & OC
  • I've filtered the data so the cards with too early or too late (the future) aren't counted at all, since if people can't fill in a date correctly, I don't trust them to know what card they bought
  • They should automatically update the data based on new submission to the main spreadsheet
I found it interesting at least as a data point to compare the unofficial calculator to just to see how different processing of the data changes how far back you might be in the queue. If anyone else is interested they can find them here: https://docs.google.com/spreadsheets/d/1L5FsPk6ZsYzN8J2XMaAoLQPbCrjKBTiCEanh1AthKs0
Click to expand...


actually it fits with my order of time and unofficial queue. i Bought my Msi Gaming Trio at 15:15 and my queue number ist 364 as your chart shows the same.
 
Status
Not open for further replies.
Back
Top Bottom