2019/02/15 09:45:39
the_Scarlet_one
I have a 1080ti and a 2070 that are chugging away, and I crossed 30 million after 14 days. Each produces about 1.05 million on average as far as I can tell. I will eventually have to stick to the monthly goals, but for now, I just let them run while the weather is cool enough for folding.

I really do like the idea of dynamically adjusting the points if needed.
2019/02/15 10:25:49
notfordman
I know we all appreciate EVGA's continued support of our folding and crunching teams. The top tier maybe a bit too high of a jump for some folders. Not everyone can afford the new cards since they have almost more than doubled in price. Not everyone can fold 24/7 either. When the weather warms up it affects some of our team mates as well. Is it possible to bring the top tier down to 20,000,000 or so? What was the considering factors to decide on these thresholds? If it was based on solely the 20xx series performance, that wouldn't be a true reflection of our team I don't feel. 
2019/02/15 11:07:39
EVGA_Lee
notfordman
I know we all appreciate EVGA's continued support of our folding and crunching teams. The top tier maybe a bit too high of a jump for some folders. Not everyone can afford the new cards since they have almost more than doubled in price. Not everyone can fold 24/7 either. When the weather warms up it affects some of our team mates as well. Is it possible to bring the top tier down to 20,000,000 or so? What was the considering factors to decide on these thresholds? If it was based on solely the 20xx series performance, that wouldn't be a true reflection of our team I don't feel. 


No, 20-Series performance alone wasn't the main driving force.  A factor, but not the only factor.


Here are a few items that were part of the consideration, with no particular order:
  • # of EVGA Bucks claimed by the Community (this will always be a factor for every promotion)
  • Performance of current Gen cards, performance of last Gen cards, performance of potential next-Gen cards, and performance of older Gen cards
  • Ease of attaining Year 11 numbers
  • Variables (one of the reasons why we're reserving the ability to change the point totals)
    • Single vs. multi card setups
      • There are rough PPD totals for most cards, but there's no easy way to determine how many folders run multiple cards that may make the numbers more attainable
    • The number of people that fold only until they hit the monthly number and stop
    • The number of people that might reconsider participating in this program after raising the point total at different levels
We certainly looked at another year of easing people into moderately higher numbers.  From my point of view, I feel this is more of a transition year to readjust the point levels to appropriate levels.  "Appropriate levels" will hopefully be determined through the year.  It might be rough for the early going, but this will setup the program better for the next few years. 
 
Again, we appreciate this feedback.  Part of the reason we announce this early is to hear from everyone.  This is one of the bigger changes we've made to the program, and it's a large task to make it correct.  We'll be reviewing this again - probably early next week - so any final comments will be appreciated.
2019/02/15 11:34:04
troy8d
EVGATech_LeeM
 
No, 20-Series performance alone wasn't the main driving force.  A factor, but not the only factor.


Here are a few items that were part of the consideration, with no particular order:
  • # of EVGA Bucks claimed by the Community (this will always be a factor for every promotion)
  • Performance of current Gen cards, performance of last Gen cards, performance of potential next-Gen cards, and performance of older Gen cards
  • Ease of attaining Year 11 numbers
  • Variables (one of the reasons why we're reserving the ability to change the point totals)
    • Single vs. multi card setups
      • There are rough PPD totals for most cards, but there's no easy way to determine how many folders run multiple cards that may make the numbers more attainable
    • The number of people that fold only until they hit the monthly number and stop
    • The number of people that might reconsider participating in this program after raising the point total at different levels
We certainly looked at another year of easing people into moderately higher numbers.  From my point of view, I feel this is more of a transition year to readjust the point levels to appropriate levels.  "Appropriate levels" will hopefully be determined through the year.  It might be rough for the early going, but this will setup the program better for the next few years. 
 
Again, we appreciate this feedback.  Part of the reason we announce this early is to hear from everyone.  This is one of the bigger changes we've made to the program, and it's a large task to make it correct.  We'll be reviewing this again - probably early next week - so any final comments will be appreciated.




I don't understand why the ease of Year 11 numbers is a factor.  Granted, the year 11 target was far too low (I advocated that Year 11 target be doubled from where it ended up).  Does this imply that you are now raising target above what they should actually be to compensate for the fact that they were low in the past?  i.e. You gave out too many in the past so now you're giving out too few so that on average you're where you want to be?  I can't see any other reason the ease of Year 11 would be relevant in setting future targets?
 
I also don't understand the relevance of future card performance.  Unless they are being released in the next few months this seems irrelevant as future points levels can be adjusted to coincide with future cards.
2019/02/15 11:39:56
troy8d
EVGATech_LeeM
 
We certainly looked at another year of easing people into moderately higher numbers.  From my point of view, I feel this is more of a transition year to readjust the point levels to appropriate levels.  "Appropriate levels" will hopefully be determined through the year.  It might be rough for the early going, but this will setup the program better for the next few years. 
 
Again, we appreciate this feedback.  Part of the reason we announce this early is to hear from everyone.  This is one of the bigger changes we've made to the program, and it's a large task to make it correct.  We'll be reviewing this again - probably early next week - so any final comments will be appreciated.

 
While I am generally supportive of a large increase in the points required, I will join the chorus of people who are grateful for EVGA's support but also concerned that the point threshold is significantly too high. 

An important issue to consider is the asymmetric nature of over-estimating the required points threshold.  Its easy to scare people away and convince them its not worth the time or effort to fold, but once they're gone its generally significantly harder to convince them to come back by lowering the threshold several months down the line.  A more conservative estimate with a gradual increase seems like a better approach than an over-estimate and gradual decrease.
2019/02/15 11:54:19
troy8d
EVGATech_LeeM
I'm happy to see a lot of constructive feedback from the community, both here and through some emails or PMs.  I think it's fair to say that there's still a narrow window for us to receive more feedback before everything's finalized for Year 12.


As someone who provided feedback via email, I'll post it here as well to weigh in with my $.01 if it furthers the debate/dialogue.
 
From the outset I want it to make it clear that my motivation is not personal and I have surpassed 25 million points monthly for the last several years (other than a mobo failure 5-6 months ago and some occasional downtime).

As someone who has long been advocating that EVGA raise the required points for both Tier 1 and Tier 2 of the folding at home program, I am very concerned that the most recent increase is far too extreme and detrimental to the EVGA folding program in general.  On the forums I had previously advocated that year 11 should have been 4 million and 8 million respectively to be in line with the power of the 10xx series GPUs and that year 12 should be 8 and 16 million to be in line with the 20xx series GPUs and stand by these numbers.
 
In the past monthly folding targets have been accessible to the casual but dedicated folder, accessible to a wide audience of potential folders that had the most current generation of GPUs.  Many of these folders typically caught the folding bug and went on to become hardcore dedicated folders. 

The numbers I proposed are in line with the historical norms from when I started folding with EVGA back in Year 3, but more importantly they are a reasonable expectation of a casual but dedicated folder.  I don't know that the explicit standard was ever set, but the increases in the first 8-10 years of the program seemed to be roughly in line with what a middle of the line xx60 GPU could fold in approximately 20-21 days.  So if someone bought a 460 (the card I started folding on) and folded 3 weeks a month they would roughly hit the tier 2 mark.  If you had a higher end GPU or a bigadv folding rig (no longer possible) you could obviously hit it sooner, but still required a substantial effort.

My best estimate of the performance of the 2060 card comes from OCN's GPU database: https://docs.google.com/spreadsheets/d/1vcVoSVtamcoGj5sFfvKF_XlvuviWWveJIg_iZ8U2bf0/pub?output=html which estimates 800,000 PPD for a 2060.  This would equal 24 million points over a 30 day month or 24.8 million points in the best case of a 31 day month, still missing target of 25 million points, while leaving no leeway to play games or allow for any (inevitable) computer downtime.  Granted, if you have a better estimate of the 2060's PPD that estimates it to be 1.25 million PPD my point is moot, unfortunately I don't have the finances to purchase a 20x0 series card to test this myself.

I fully agree that last year's numbers were far too low, and a substantial increase is warranted.  I am concerned that they are too extreme and anyone without a 2070 card or better (or multiple GPUs) will not be able to hit them.  Upgrading to the current generation of GPUs to hit the latest tiers is to be expected, but upgrading to a higher power card and folding longer is unprecedented.  When you combine this with significant increase in price of GPUs over the past 10 years and the erosion of the US dollar (and correspondingly EVGA buck) due to inflation over the past 10 years, the current proposal for year 12 appears very detrimental to a causal but dedicated folder in the EVGA folding community.

A casual folder now has to buy the xx70 card rather than the xx60 card, which is more expensive, and has to fold on it as long or longer than was expected in the historical norm (excluding the most recent few years) only to be rewarded less seems very detrimental to the EVGA's support for the Folding at Home program and the folding community at EVGA.

I don’t dispute the need to increase the points required to hit tier 2 in Year 12, but more importantly I would urge them to do so in a manner that is not so exclusive.  Keep in mind that most folders spend far more on electricity than they do on hardware to support this cause, and most people fold because its a good cause worth supporting, not simply for the EVGA bucks (though they are a nice bonus and we are grateful for the support and also recognize that some simply are bucks folders).  I would also urge EVGA to increase the reward to 12 or 13 EVGA bucks to keep pace with inflation, but that concern is small in comparison to keeping the tiers within reach of a casual but dedicated folder.

Thank you for your time and for continuing your support for the EVGA Folding at Home program.  The last thing I want to do is come across as entitled or ungrateful, but I also want to see the program continue to thrive and be accessible to all folders.
2019/02/15 12:01:31
EVGA_Lee
troy8d
I don't understand why the ease of Year 11 numbers is a factor.  Granted, the year 11 target was far too low (I advocated that Year 11 target be doubled from where it ended up).  Does this imply that you are now raising target above what they should actually be to compensate for the fact that they were low in the past?  i.e. You gave out too many in the past so now you're giving out too few so that on average you're where you want to be?  I can't see any other reason the ease of Year 11 would be relevant in setting future targets?
 
I also don't understand the relevance of future card performance.  Unless they are being released in the next few months this seems irrelevant as future points levels can be adjusted to coincide with future cards.

It's probably more accurate to say Ease/Difficulty of Year X numbers. Certainly we'll review the same after Year 12, when we go over the numbers for Year 13. Although this factor in and of itself may not be the most important factor, it's heavily related to some of the other factors on the list. For example, if people can reach both tiers within a week with current and much older hardware, then we aren't setting the right target numbers.  Conversely, if we see a lot of drop-off after the year, then we'll need to revise the targets.
 
Overall, it's more of a factor this year simply because the performance creep of cards show that Year 11 numbers have finally hit a point where the numbers are far out of touch with current performance capabilities of cards for the high-end and lower-end hardware.  The intention is to not rubber band the opposite direction, despite its appearance, but we do need to set higher goals regardless.
2019/02/15 12:05:47
troy8d
That makes much more sense.  Year 11 was certainly an outlier that was far too low of a target.  Thanks for clearing that up
2019/02/15 12:08:09
EVGA_Lee
troy8d
EVGATech_LeeM
I'm happy to see a lot of constructive feedback from the community, both here and through some emails or PMs.  I think it's fair to say that there's still a narrow window for us to receive more feedback before everything's finalized for Year 12.


As someone who provided feedback via email, I'll post it here as well to weigh in with my $.01 if it furthers the debate/dialogue.
 
[...]

Your email was forwarded to me this morning, and it's one of the reasons why I'm carefully reading much of the feedback the community is giving on this.  Again, I greatly appreciate all the thoughtful responses.  You guys (the EVGA F@H team) care very much about not only the cause, but the health of the team, so I'm paying close attention to a lot of these responses.
2019/02/15 12:14:18
troy8d
EVGATech_LeeM
Your email was forwarded to me this morning, and it's one of the reasons why I'm carefully reading much of the feedback the community is giving on this.  Again, I greatly appreciate all the thoughtful responses.  You guys (the EVGA F@H team) care very much about not only the cause, but the health of the team, so I'm paying close attention to a lot of these responses.



I can assure you that the feeling is mutual.  The folding community greatly appreciates all the time, effort and support that EVGA dedicates to the folding at home team.  I (and many others) would not bother providing feedback if we felt it would be wholly ignored. 

Use My Existing Forum Account

Use My Social Media Account