Discrete OpAmp Review: Sonic Imagery vs Sparkos

[rubley00]

Nothing was missed in the measurements. Measurements showed a difference. Question is how audible it is.

I'm not claiming you missed anything, I feel your tests capture what is important. I also believe anything that could be heard can be measured. If errors show up in the measurements, performing a subjective test cannot be used to confirm the measurements.

By the link you referenced, there is no question about how audible it is. There are charts for that. And what is audible to you may not be to someone else or vice versa, making your subjective evaluation moot.

It would be like measuring the temperate of water at 10 deg C, and then telling me you confirmed that the water isn't frozen.

 

[rubley00]

LCD isn't, unless you hate contrast and low latency. Only OLED and microLED technologies can help us reach the future SED/FED was supposed to be.
And I say that as someone with an expensive Eizo monitor.

As someone who writes code all day, and I need to see as much on the screen as possible at once, I was happy to ditch a 20" CRT with corner convergence issues to get an early 20" LCD with perfectly crisp pixels. Contrast ratio is a trade off and early LCDs weren't great, but the clarity was something CRTs never achieved.

 

[Nathan Raymond]

LCD and OLED as technologies are a massive step forward. The only argument to be made is that modern TVs don't do a very good job scaling video signals from 25 year old game consoles. Perhaps some games were also coded in a way that presented the visuals in a certain way on a CRT, but that's not an indictment of LCD/OLED.

For LCDs, IPS panels have great viewing angles but poor contrast and so-so response times, TN panels have poor color range (often using temporal dithering to simulate 8-bits per channel) and poor viewing angles but good response time, while VA panels have very good contrast ratios but compromise on the color gamut and viewing angles while being ok on response time. Proper HDR for LCDs is achieved by an array of individually addressible backlight LEDs and additional image processing by the panel to adjust the brightness of those zones appropriately. Choosing a good LCD is about choosing your poison. I go with VA panels because static contrast rations less than 2500:1 look bad and I can deal with the other VA downsides. And let's talk refresh rates for a moment. While CRTs can do anything within a range no problem, LCDs are traditionally very limited and have trouble coping with anything outside of a narrow range. Only until the last few years have we gotten NVIDIA's proprietary G-Sync (which requires an additional expensive processing board to be installed/bundled in the display enclosure and only works with NVIDIA's video cards) or AMD's FreeSync which is now part of the latest HDMI 2.1 Variable Refresh Rate (VRR) standard. Previously, if you cared about movies like I and others do, and you wanted to see 24p playback (i.e. no judder) you had to get a 120Hz or other refresh multiple of 24 TV/display that could do an integer pulldown like 5:5 on a 24p incoming signal. On pre-VRR TVs, you have to deal with the overhead of refresh rate changes (on my Roku every time it goes from a menu to playing back video or back again and adjusts the refresh rate my LG LCD goes blank while it adjusts which drops the HDMI connection long enough that it triggers the Roku to throw up a big warning screen about HDMI bandwidth which goes away after a second or two once the LG TV finishes adjusting - which would never be a problem on a CRT). Sure LCD/OLED have some advantages over CRT, but in the big picture, it's all sets of tradeoffs. The one big thing I'll give LCD and OLED is that they are physically a lot lighter and take up less space.

 

[rubley00]

which drops the HDMI connection

Yes, HDMI sucks, there are high end receivers that maintain the negotiation between devices so it doesn't drop, but yeah it sucks.

LCD/OLED has nice crisp pixels and avoids the convergence issues of CRT, which was a godsend for desktop monitors.

 

[q3cpma]

As someone who writes code all day, and I need to see as much on the screen as possible at once, I was happy to ditch a 20" CRT with corner convergence issues to get an early 20" LCD with perfectly crisp pixels. Contrast ratio is a trade off and early LCDs weren't great, but the clarity was something CRTs never achieved.

As a programer, I agree about that part (especially with bitmap fonts), but for multimedia, good CRTs (like Sony GDM-FW900) were a lot better than IPS/PLS/AHVA. And let's not forget that we got TN at the beginning.
Also, flat CRTs exist.

 

[q3cpma]

Previously, if you cared about movies like I and others do, and you wanted to see 24p playback (i.e. no judder) you had to get a 120Hz or other refresh multiple of 24 TV/display that could do an integer pulldown like 5:5 on a 24p incoming signal.

This problem was solved some time ago by madvr's smoothmotion, then by mpv's display resampling (https://github.com/mpv-player/mpv/wiki/Interpolation).

 

[rubley00]

As a programer, I agree about that part (especially with bitmap fonts), but for multimedia, good CRTs (like Sony GDM-FW900) were a lot better than IPS/PLS/AHVA. And let's not forget that we got TN at the beginning.

The first LCD's weren't great, I had the first 20" LCD Dell created, paid $1000 for it, and happily dumped the 19" Sony I had that I had also paid ~$1000 for. Edge-to-edge readability was worth the loss of contrast and colors.

 

[q3cpma]

The first LCD's weren't great, I had the first 20" LCD Dell created, paid $1000 for it, and happily dumped the 19" Sony I had that I had also paid ~$1000 for. Edge-to-edge readability was worth the loss of contrast and colors.

What I (or we) mean is that's it's not "a big step forward", only a different compromise. Which is why microLED is incredibly exciting, since it's the end of compromises.

 

[rubley00]

What I (or we) mean is that's it's not "a big step forward", only a different compromise. Which is why microLED is incredibly exciting, since it's the end of compromises.

I included OLED in my statement. You're going to have a hard time convincing me OLED is marginal improvement to CRT. A CRT could never achieve perfect RGB alignment at 3840x2160. CRTs couldn't do that at 1600x1200.

 

[Nathan Raymond]

I included OLED in my statement. You're going to have a hard time convincing me OLED is marginal improvement to CRT. A CRT could never achieve perfect RGB alignment at 3840x2160. CRTs couldn't do that at 1600x1200.

I like most of the things about OLED, but the operational lifespan of the blue elements is a major drawback and why I don't consider OLED a viable option. Fingers crossed microLED delivers on its promises, since it doesn't have that problem.

 

[tensor9]

I like most of the things about OLED, but the operational lifespan of the blue elements is a major drawback and why I don't consider OLED a viable option. Fingers crossed microLED delivers on its promises, since it doesn't have that problem.

They’ve sort of engineered the problem away by automatically adjusting the voltage applied to the blue subpixel over time, so that the burghness stays constant. All said; the lifetime is still plenty good enough for the typical lifespan of a TV.

 

[pozz]

This is a killer Christmas season. I have spent more money this year than any other and by far.

So I appreciate you all donating generously using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Hidden reference to the Klippel NFS.

 

[JohnYang1997]

Recently I made a discrete opamp.

 

[Veri]

Recently I made a discrete opamp.

Lol you are into everything!

is it any better than Sonic Imagery, Sparkos? (any benefit compared to traditional op-amps? I'm expecting no?)

 

[JohnYang1997]

Lol you are into everything!

is it any better than Sonic Imagery, Sparkos?

According to measurements I have seen so far. It's way better. Dunno if Bruno has made something better. And I didn't even use precision matched transistors.

 

[March Audio]

View attachment 64391Recently I made a discrete opamp.

Cool!

What's the thd + n V something like an OPA1612?

 

[Veri]

According to measurements I have seen so far. It's way better. Dunno if Bruno has made something better. And I didn't even use precision matched transistors.

If Topping does not do anything with it, try and sell your design to Burson lol

 

[JohnYang1997]

Cool!

What's the thd + n V something like an OPA1612?

AP can't measure the difference in distortion and noise. Except for the discrete can have higher supply voltage. The number will be unlimited to 0.00005x%.

 

[JohnYang1997]

Cool!

What's the thd + n V something like an OPA1612?

Do you still need the graph tho?

 

[March Audio]

Do you still need the graph tho?

I'm just interested to see an overall comparison - not limited to this specific parameter.