# Flame Benchmark Archive
**URL:** https://forum.logik.tv/t/flame-benchmark-archive/987
**Category:** Flame Questions
**Created:** [October 5, 2020, 2:50pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987 "2020-10-05T14:50:18Z")
**Posts on this page:** 20
**Page:** 2
### Author:  [@ALan](https://forum.logik.tv/u/ALan)
#### Post date: [April 20, 2023, 9:24pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/22 "2023-04-20T21:24:42Z")
We do all day and night long.
> [@space\_monkey](#):
>
> working in a facility off a server is the biggest hurdle to overcome imho, the servers aren’t fast enough for say 5 flames to be able to playback
---
### Author:  [@ALan](https://forum.logik.tv/u/ALan)
#### Post date: [April 20, 2023, 9:26pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/23 "2023-04-20T21:26:43Z")
Exactly. But the reality is we save hours a day.
> [@Ryland](#):
>
> I agree that its definitely a risk, but people are vastly underestimating the time lost daily to a cached workflow. You only need to save a few minutes a day to make up for the occasional downed server.
---
### Author:  [@ALan](https://forum.logik.tv/u/ALan)
#### Post date: [April 20, 2023, 9:45pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/24 "2023-04-20T21:45:05Z")
Given that in such an environment, each workstation already has a high speed fiber connect to the server, it is very likely not much additional cost. Once you have 25GBe network, that easily covers 4K+ resolutions. Large NVMe SSD drives are super cheap now. 15TB for around ~1800. 4TB SATA SSD is $260. Easy to build a 200TB SSD server for under $30k. With 15 flame workstations, you’d make back the cost with effeciency in less than a month easy. Depending on how you calculate, could even be a week.
> [@space\_monkey](#):
>
> sure but the cost of the hardware to let say 15 flame artists being able to playback 4k without hiccups
> would be more expensive I am guessing than the cached workflow in terms of investment
---
### Author:  [@space\_monkey](https://forum.logik.tv/u/space_monkey)
#### Post date: [April 21, 2023, 2:39am UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/25 "2023-04-21T02:39:57Z")
some of the bigger facilities could have around 90 artists working at once clearly that would slow playback even on a 25 gbe line some doing 8k I cant imagine that would handle that throughput but I dont disagree with you it clearly has its place .
---
### Author:  [@ALan](https://forum.logik.tv/u/ALan)
#### Post date: [April 21, 2023, 5:52am UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/26 "2023-04-21T05:52:36Z")
25GBe on the workstation. Server side could easily be 100/200/400 gigE. I don’t know of any facility with 90 Flames active at a single within a single location.
> [@space\_monkey](#):
>
> some of the bigger facilities could have around 90 artists working at once clearly that would slow playback even on a 25 gbe line some doing 8k I cant imagine that would handle that throughput but I dont disagree with you it clearly has its place .
---
### Author:  [@space\_monkey](https://forum.logik.tv/u/space_monkey)
#### Post date: [April 21, 2023, 1:25pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/27 "2023-04-21T13:25:32Z")
Cool, thanks for the info.
---
### Author:  [@space\_monkey](https://forum.logik.tv/u/space_monkey)
#### Post date: [April 24, 2023, 12:15am UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/28 "2023-04-24T00:15:11Z")
@ALan I understand your perspective. However, convincing a corporation to invest in a solution that pays off in just a month can be a challenging task, especially when the required investment is just a few NVME storage solutions for a few hundred dollars. The politics involved can be overwhelming, and pushing the solution through the corporate structure can be difficult. Additionally, the real challenge is how Flame archives its material. If Flame were to retool its archiving mechanism, it could potentially solve the problem with a software solution, making it irrelevant how an artist or corporation decides to work. For instance, if archiving were set up to gather source material and archive it uncached with direct links back to the source material, or using other scenarios, storage costs would not be an issue, and the size of the archive and the difficulty in dealing with it would be reduced.
---
### Author:  [@Nick](https://forum.logik.tv/u/Nick)
#### Post date: [May 15, 2023, 7:53pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/29 "2023-05-15T19:53:15Z")
Was just wondering if anyone could explain the difference between frame based / track based rendering. Frame is obviously faster. Why the choice?
---
### Author:  [@AdamArcher](https://forum.logik.tv/u/AdamArcher)
#### Post date: [May 15, 2023, 8:35pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/30 "2023-05-15T20:35:20Z")
> [@ALan](#):
>
> We do all day and night long.
Same here. At my current facility and previous.
---
### Author:  [@Jerem](https://forum.logik.tv/u/Jerem)
#### Post date: [May 15, 2023, 11:21pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/31 "2023-05-15T23:21:37Z")
The difference is when you select all your layer and presse render.
If it’s set to track based, it first calculate the entire track and then Switch to the upper one.
Frame based is calculating layer 1 frame 1 / layer 2 frame 1…
You should go for track base render, I’ve done many test and most of the time it’s faster and you might have less crash.
---
### Author:  [@randy](https://forum.logik.tv/u/randy)
#### Post date: [July 7, 2023, 1:56pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/34 "2023-07-07T13:56:56Z")
Weird. They are in fact missing. @Frank mind taking a leak when you have a sec?
---
### Author:  [@randy](https://forum.logik.tv/u/randy)
#### Post date: [July 7, 2023, 4:20pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/36 "2023-07-07T16:20:39Z")
Well I did ask nicely. :-). Hahahahahahahaha
---
### Author:  [@hBomb42](https://forum.logik.tv/u/hBomb42)
#### Post date: [August 18, 2023, 3:14pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/37 "2023-08-18T15:14:04Z")
Hi all! Conversations are starting up at the office about what to replace our aging z840s with. I’m staunchly on Team Linux, but would love to see updated data. If there are any folks bored out there on Friday afternoon with 2024.1 and M2 Studios or beastly threadripper setups that wanted to add their results to the Sheet, it’d do me…I mean “the community” a solid!
> **[Flame 2020 4K Benchmark times & specs](https://docs.google.com/spreadsheets/d/1Wc9U_5AsRmSXM3DZAXh0H4fIDgDOJjJT2MwWKLeUa9Q/edit#gid=0)**
>
> Sheet1
>
> ,Flame 2020 archive can be downloaded from this link
> FrankiE Cueto,2023-04-27,Flame 2024,RTX 4090 FE,3:04:00,Lenovo P620 AMD Ryzen Threadripper Pro, 3975WX 3.5Ghz, 128GB RAM, 2TB NVME,Rocky 8.7
---
### Author:  [@randy](https://forum.logik.tv/u/randy)
#### Post date: [August 18, 2023, 6:06pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/38 "2023-08-18T18:06:59Z")
If maximum return on your investment is important, don’t discount the upgradability and repairability of Linux.
---
### Author:  [@hBomb42](https://forum.logik.tv/u/hBomb42)
#### Post date: [August 18, 2023, 6:31pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/39 "2023-08-18T18:31:49Z")
I suspect the battle I’m wading into is primarily based on support requirements, but yeah…I know all that jazz and am firmly on team linux. I’m just curious how things look in our brave new Vulkan world.
---
### Author:  [@Waldi](https://forum.logik.tv/u/Waldi)
#### Post date: [November 20, 2023, 5:35pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/40 "2023-11-20T17:35:50Z")
Would be nice to see an M3 Max and it’s values here at the Flame Benchmark list. If an early bird could provide data - that would be very appreciated.
---
### Author:  [@Sinan](https://forum.logik.tv/u/Sinan)
#### Post date: [November 20, 2023, 8:45pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/41 "2023-11-20T20:45:27Z")
Well I had done the test but completely forgot to fill in the database. I think it was around 5:20 on the new MacStudio top of the line, fully loaded cpu+ram. I’ll test it again tomorrow and report…
---
### Author:  [@Waldi](https://forum.logik.tv/u/Waldi)
#### Post date: [November 20, 2023, 8:58pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/42 "2023-11-20T20:58:15Z")
Sinan - cok merci. But I was really interested into the new M3 Max with at least 64gb RAM. These silicons are currently avail within the new MacBook Pros only.
---
### Author:  [@snacks](https://forum.logik.tv/u/snacks)
#### Post date: [November 20, 2023, 9:25pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/43 "2023-11-20T21:25:39Z")
@Waldi I should have mine with 128gb ram in my hands in two weeks and can run it then.
---
### Author:  [@Sinan](https://forum.logik.tv/u/Sinan)
#### Post date: [November 20, 2023, 9:37pm UTC](https://forum.logik.tv/t/flame-benchmark-archive/987/44 "2023-11-20T21:37:12Z")
Oh right. Yeah wrong config. Though I’m looking forward to the M3 as well…
[Previous page](https://forum.logik.tv/t/flame-benchmark-archive/987.md?page=1)
[Next page](https://forum.logik.tv/t/flame-benchmark-archive/987.md?page=3)