Thread for LTT Labs Test Suggestions

[Anfros]

When you test headphones or speakers etc it would be really cool if you'd publish compensatory eq curves compared to your target. And maybe also evaluate products with the eq so we get an idea of whether the headphones/laptop speakers/etc are bad or if it's just bad eq.

[Agall]

RTX 3000	CUDA Cores	%cores	%cores	CUDA Cores	RTX 4000
GA102	10752	100%	100%	18432	AD102
3090ti 24GB $2000	10752	100%
3090 24GB $1500	10496	97.6%
3080ti 12GB $1200	10240	95.2%
			88.9%	16384	4090 24GB $1600
3080 12GB $800	8960	83.3%
3080 10GB $700	8704	81.0%
3070ti 8GB $600	6144	57.1%
3070 8GB $500	5888	54.8%	52.8%	9728	4080 16GB $1200
3060ti 8GB $400	4864	45.2%	41.7%	7680	4070ti 12GB $800
3060 12GB $330	3584	33.3%	31.9%	5888	4070 12GB $600
3050 8GB $250	2560	23.8%	23.6%	4352	4060ti 8GB $400
			20.8%	3840	4060 8GB $300

My suggestion would be to augment this matrix with markbench scores (if I'm remembering the name of y'all's benchmarking tool properly). 

 

LTT video could tagline on some more Nvidia shaming since if the architectural/lithographic performance between Ampere and Ada that I've seen of 85-95% when adjusting for GPU binning is true, the 20-40% we're getting in improvements by name is quite disgusting (on top of the higher price for a respective tier of GPU).

 

We even have a sample of each with exactly 5888 CUDA cores, so the gap between those (architectural improvements per core) multiplied by (18432/10752) (difference in lithography) and it should give a quite accurate intergenerational multiplier from a single sample of card. Same with comparing the RTX 3070 to RTX 4080 since they're 'close enough' in %cores, which can be adjusted with a 1.038 multiplier. It shouldn't be a perfect 1:1 in CUDA core count as a multiplier, but it would be interesting to see what that is practically.

 

 

[joebroke]

I'd love to see a review and testing of the new Festool Hearing protection GHS 25 I .

 

How well do they isolate noise but how do they sound?  Are they hot garbage compared to other ear buds?

 

https://www.festoolusa.com/products/new-tools/new-tools/577793---ghs-25-i-us?sc_src=email_2832784&sc_lid=204399835&sc_uid=bPN73JkXPK&sc_llid=305591&sc_customer=0039W000002CjtYQAS#Overview

 

Thanks!

[LogicalDrm]

On 6/6/2023 at 9:59 PM, Evi said:

Labs testing the new M2 Mac Pro

 

With the new M2 Mac Pros soon coming out, I think the labs team should add transcoding to be a part of their testing. Different formats from different cameras, anything from open gate arriraw to proress 4444 XQ transcoding to Prores Proxy and Dnx36.
 

Macs are heavily optimized to work with prores, it would be cool to see them face off with comparably priced PCs.

While resolve can't be used to transcode to prores on PC, Scratch Play Pro is actually licensed by Apple to transcode to native prores - nothing hacky about it.

While editors in film might prefer PCs, the pipeline from camera to the editor is dominated by Macs. Transcode speeds in frames per second aren't something that Apple (or Intel, AMD, or Nvidia for that matter) advertise or even mention anywhere and the only tests available can be found in tight-knit DIT forums. This could be infinitely useful for us, thanks.

Merged to test suggestions thread.

[Samueras]

Hey you where thinking about a Review system where only trusted rewiever could post. I would suggest a different route. Make it so that everyone with a Account can Post reviews. Then enable either the staff or community or everybody vote on if a Review is help full neutral or harmful. Now you can add a Filter on the Website to only include reviews from users with a certain standing or higher for the rating. And make the Standart Filter resonable high.

 

That way new rewiewer can enter the frai and even their first Review  is going to have weight as soon as they got prpven themselves. And they can prove themselves from the Start on.

 

Also if a product doesn't have ratings from proven reviewers, visitors could lower the filter to maybe still find a Review.

[ollioddi]

I am not sure if this is the right place, or if this is even on the drawing board. I am watching the 4060 review and as you mentioned, you have done an absolutely huge amount of tests for this card. Also, it's obvious that the card performs like garbage.

 

In regards to this, do you have plans for a hardware comparison site with all of the data you collect?

 

A feature that would be awesome to have, would be for the user to enter the local price for cards, and based on your performance measurements, calculate the best performance per dollar. I do believe that this would be surprising in many cases, where they don't perform as expected.

 

Lets say im looking to buy the 4060. The site could perhaps suggest surrounding cards, and ask the user to enter their price for each card. That would be such a neat feature, since price can vary alot depending on location.

 

 

[LMGcommunity]

3 hours ago, ollioddi said:

I am not sure if this is the right place, or if this is even on the drawing board. I am watching the 4060 review and as you mentioned, you have done an absolutely huge amount of tests for this card. Also, it's obvious that the card performs like garbage.

 

In regards to this, do you have plans for a hardware comparison site with all of the data you collect?

 

A feature that would be awesome to have, would be for the user to enter the local price for cards, and based on your performance measurements, calculate the best performance per dollar. I do believe that this would be surprising in many cases, where they don't perform as expected.

 

Lets say im looking to buy the 4060. The site could perhaps suggest surrounding cards, and ask the user to enter their price for each card. That would be such a neat feature, since price can vary alot depending on location.

 

 

A Labs site is in the works!

[10leej]

I would like to see an FPS per watt comparison across GPU's as well as benchmarks for x264, x265, VP9, and av1 encoding/decoding. I know you've dealt with render tests in the past but it's criminally difficult to find data on this as a non gamer. When I buy a new GPU there's NO FREAKING benchmarks just anecdotal evidence from comments in a htpc forum and usually that just boils down to "buy nvidia"

 

of which I wanna know, just how good is the hardware acceleration for Intel Arc's Quicksync? How improved is AMD's AMF?

 

Can we get more Compute benchmarks for GPU's as well? maybe a actual fucking chart for what AMD GPU's suppoer RoCM?

[LogicalDrm]

On 6/28/2023 at 12:13 PM, Samueras said:

Hey you where thinking about a Review system where only trusted rewiever could post. I would suggest a different route. Make it so that everyone with a Account can Post reviews. Then enable either the staff or community or everybody vote on if a Review is help full neutral or harmful. Now you can add a Filter on the Website to only include reviews from users with a certain standing or higher for the rating. And make the Standart Filter resonable high.

 

That way new rewiewer can enter the frai and even their first Review  is going to have weight as soon as they got prpven themselves. And they can prove themselves from the Start on.

 

Also if a product doesn't have ratings from proven reviewers, visitors could lower the filter to maybe still find a Review.

Merged to Labs suggestions thread. Using external data sources was also discussed here. Your suggestions is more in line of PCPP review system or other user reviews on several long-running forums like ours, Tom's etc.

[JuJuTheWulfPup]

On 5/20/2023 at 9:05 AM, yesiamzach said:

- headphone ear cup depth, I don't like when my ear touches the inside of my headphones, again my QC35ii's fit my ears without touching, the sony WH-1000XM2's do not, they push my ears back towards the side of my head.

I second this, I'm trying to find replacement ear cushions (aka ear pads or ear cups) for the HyperX Cloud Mix, and learned later on that I cannot share this headset with my husband because the ear cup space (inside of the ear pad) is too small for him. I can't seem to find any website that lists:

• inside ear pad measurements (height, width, depth -- to determine ear fit), and
• outside ear pad measurements (height, width, thickness) and perhaps diameter of headphone without the ear pad attached (to determine ear pad fit)

I've been frustrated with this for over a year as the stock leatherette pads for my HyperX Cloud Mix are crumbling to pieces and sticking to my face, but only recently learned that the Cloud Mix isn't the only smaller HyperX headset, but I can't seem to find any information about the ear pad/cushion, or ear cup size of the Cloud Alpha S to figure out if the Cloud Mix is close in size or not.

For anyone that wanted to help in this search and research, feel free to DM me. But I would really prefer to not buy ear pads that will fit a normal-sized HyperX Cloud. (I have tested the cushions from my old Cloud Stinger and my it does wiggle around, so idk whether I should be willing to buy ones that fit the majority of Clouds or not.)

[jadsa99]

The photography/videography community would very much appreciate some in-depth tests on the various brands of CFExpress Type B cards. Currently, there aren't many options that work well for 8k footage recording in raw formats, and the number of different places I've had to look to try and get a comprehensive list is ridiculous. (A good example of what i'm talking about) One of the biggest gripes that I (and others) have is that many of the offerings advertise speeds that can support the high bitrate video, but fail at the task of sustained writes. With more and more cameras being released into the consumer/prosumer segment that require those high sustained write speeds, I think it could be a very valuable way to solidify the lab as the place to go when researching which Type B drives are worth their price.

 

No hate to PetaPixel, they have a great breakdown and was one of the best tests I was able to find, but it was so far from comprehensive, even after being updated, that I found myself searching for hours to check out the performance for some of the other brands that are available.

 

Lastly, there is an increasing number of people (myself included) that have been intrigued/have tried out making their own Type B cards via simple adapters like this that take a 2230 nVME drive and adapt it for use as a CFexpress Type B. They offer a much better price: performance ratio if they work as advertised.

 

Like I said, there are a ton of different overviews of the different cards, readers, and accessories possible, but I have yet to find anything comprehensive enough to feel like I'm making the right purchase for my camera.

 

 

[andresvilas]

New Mac Pro M2: virtualize  8 Mac Minis 

[IanSzot]

I'd love to see Labs testing if these low latency guide for Windows gaming make any difference at all in modern gaming with modern hardware, might even fit for a LTT video too.

 

https://github.com/BoringBoredom/PC-Optimization-Hub

 

https://docs.google.com/document/d/1c2-lUJq74wuYK1WrA_bIvgb89dUN0sj8-hO3vqmrau4/edit?pli=1

[BardTheTrue]

I'd love for LTT to start sharing their LTT Lab data on their website in the form of Web API's. Think of the amount of tools developers can create for consumers to help avoid bad or unsafe purchases.

Just two examples I've thought of: PCPartPicker could connect to the API's for these power supply data and provide a warning for or disable selecting unsafe power supplies. Someone to could create an app that compares their GPU benchmark scores to current prices on various retailers to get New or Used Price/Performance Gaming or Productivity Value Score.

As a Software Developer I think they have a huge opportunity for good in this space here. What other ideas do you guys think this data could provide?

[LogicalDrm]

13 hours ago, BardTheTrue said:

I'd love for LTT to start sharing their LTT Lab data on their website in the form of Web API's. Think of the amount of tools developers can create for consumers to help avoid bad or unsafe purchases.

Just two examples I've thought of: PCPartPicker could connect to the API's for these power supply data and provide a warning for or disable selecting unsafe power supplies. Someone to could create an app that compares their GPU benchmark scores to current prices on various retailers to get New or Used Price/Performance Gaming or Productivity Value Score.

As a Software Developer I think they have a huge opportunity for good in this space here. What other ideas do you guys think this data could provide?

Merged to Labs suggestions thread.

[xSPOONx]

Hi LTT,

Just finished watching the video about the PSUs from Yesterday, July 11. Something you may consider adding, that might also add another layer of testing/data is a TVSS. I'm a Red Seal Electrician and if you start down the rabbit hole of harmonics, which was indirectly mentioned in the video, you're entering a not well-understood area of mystery - but it can have a significant impact on the grid and local devices. I have seen Powerline Ethernet Adaptors work flawlessly at multiple locations over various differences, but then not work at all between 2 directly connected receptacles for example. You can also drastically shorten the life of a product due to dirty power, but without expensive equipment, you'd never know there's a problem. They never used to be a problem because very few things made use of products that created harmonics, but in our digital world today,  nearly everything has the potential to create them as well as messing up power factor and efficiency. 

I've attended training on TVSS systems and a product called CleanVolt (a Canadian company based in Calgary), and some seem to work and do something. It can get super technical and may not make for the best videos, but it might be data worth checking out. You could also check out the efficiency of all the different LED lights and other electronics used on set or at home. I doubt that EVGA PSU is a one-off. I have actually heard that because of all the changes that happen to take Solar and  Wind from DC input to  AC  output things can get really messy because the converters they're using aren't always the best and they'll occasionally end up with something like that EVGA PSU. The frequency or power factor can be thrown off,  which leads to a significant hydro benefit being that once it's set up and running the turbines will spin at a fixed speed to generate their potential at exactly 60Hz without deviation. With wind and solar, you're forced to rely on conversion which can be really good or really lousy.

[Cookiebot]

Having recently navigated the gaming mouse market, I've noticed a significant gap in readily available information about mouse software / driver capabilities. I propose that LTT Labs conduct thorough testing and reviews of gaming mouse software as part of your lab's test suite.

My main pain points were:

• Difficulty in identifying which software supports a DPI clutch ('sniper') button.
• Comparing the macro capabilities between different software (for example possibility to add a random delay to key events).
• Understanding the user interface and usability of each software.

I believe these tests would greatly benefit our community. Here's why:

1. Informed Decision Making: Detailed reviews would provide the community with critical insights into software capabilities, thus enabling informed buying decisions.

2. Time-saving: By offering a one-stop resource for software functionalities, we can save community members countless hours spent researching obscure Reddit threads and YouTube videos.

3. Bridging the Knowledge Gap: Many gaming enthusiasts focus on hardware specs, while software remains an underexplored area. LTT Labs has the chance to fill this gap and guide users on both fronts.

Thank you for considering this proposal. I'm confident that it will significantly enhance the value of our community's discussions and user experiences.

[BaidDSB]

Please review budget TV's like TCL C series and T6G

 

https://www.amazon.in/TCL-inches-Ultra-Google-55T6G/dp/B0C2VJMKZS/ref=sr_1_4?crid=1VHKDZIW77I5N&keywords=tcl+4k+qled&qid=1689337516&sprefix=tcl+4k+qle%2Caps%2C216&sr=8-4

 

https://www.amazon.in/TCL-Certified-Android-55C825-Graphite/dp/B093WNSYRP/ref=sr_1_5?crid=1VHKDZIW77I5N&keywords=tcl+4k+qled&qid=1689337516&sprefix=tcl+4k+qle%2Caps%2C216&sr=8-5

[LeapFrogMasterRace]

I got a video that NEEDS to be made. Perfect to break in the new labs team.

 

So I have a 5800x with a MSI B550 that has USB disconnect issues. No big deal right just RMA the board? Well millions of people have this issue too. I will include screen shots below. AMD allegedly patched this with a BIOS update AGESA patch 1.2.0.3b. Now this did fix it for some people. Mostly only for light users i.e keboard/mouse 1 controller that's it. Even after this there are still countless people with the issue. So I thought about switching to intel LGA 1700, turns out the issue exist there too. To replicate this issue you need to heavily use the usb ports i.e large file transfers, several devices connected etc. You will still get random disconnects without high USB port usage however it is not common.

 

AMD was asked directly about what caused this issue on their products, they outright refused to answer. Perhaps to avoid a recall?

 

This appears to be related to all these rapid changes being made to modern chipsets, perhaps there is some serious USB compatibility issues going on. 

 

Take a look at how prevalent this issue is. 

 

z690

AMD Side (Please ignore the AMD fanboys saying its fixed with AGESA 1.2.0.3b. Its not they are probably very light users of USB ports and not power users. Yes 18 MILLION results 

 

 

 

This needs to be investigated as USB ports are much appreciated I can provide any additional information or documentation if requested 

[Niko-O]

Can we expect to see the recently released Framework 16 being tested? I'm considering to buy one (since I'm fed up with problems from my current laptop) but I'd love to have some real-world figures before making that decision.

(I'm actually surprised there wasn't a video yet.)

[saintlouisbagels]

9 minutes ago, Niko-O said:

Can we expect to see the recently released Framework 16 being tested? I'm considering to buy one (since I'm fed up with problems from my current laptop) but I'd love to have some real-world figures before making that decision.

(I'm actually surprised there wasn't a video yet.)

It's not released. Framework is currently taking pre-orders.
I believe it's shipping Q4 2023.

[Niko-O]

2 minutes ago, saintlouisbagels said:

It's not released. Framework is currently taking pre-orders.

Ah, yes, that's correct. But considering Linus' stance with Framework, I'd expect them to provide him one for testing (even if it might be relatively early).

[ErrorNull_]

It would be great for smaller AC to USB adapters if a noise profile could be added as well. Since I've noticed some cheaper adapters emit a very high-pitched hissing noise especially when using a lot of power, which is a total deal breaker for me.

[EggOfGlory999]

Try to test this build to see if you can do any gaming on it!

The WORST 64-Bit PC(?)

[r0ckstar]

Any chance the labs can test a bunch of baby monitors and give us a true review. Search online can't even return a true review. I've tryed so many and they are all garbage. Motorola was just basic with no features but kinda worked, vtech had nice features but couldn't stay connected to a wifi AP for the life of it even if you taped it to the AP.

 

Just seems like a market that is filled with rebranded junk or super old technology.

 

Maybe even see if Jake or someone can build and do a video of making your own using a IP camera and home assistant or something like that