Siglent SSA3000X and SSA3000X-Plus Spectrum Analyzers

[Wuerstchenhund]

Here  waterfall image how my SSA3000X  display it. It show level only with color.
(I can not test without this AMK-SSA3000X Option because permanent Option can not remove using SSA GUI)
This looks very different what PC software EasySpectrum image @Bicurico  posted  and it looks 3-d waterfall.

Stricly speaking only the image from your SSA-X shows a waterfall display format (the EasySpectrum image from Bicurico isn't a "waterfall" display format, it's just a 3D P/F/t display format; waterfall display formats, as the name implies, start at the top and built up to the bottom, with the vertical axis being the time axis).

(note also image quality when used JPG or PNG where PNG wins like 10-1.  I hope Siglent add this PNG and make it as default. Peoples can take selfies and scenery or what ever art pictures using JPG but this is not suitable for this kind of technical document images.

That is not correct. JPEG can provide a similar image quality as PNG, but it depends on the parameters (i.e. quality level, compression factor) that are used for generating them. If the screenshots in JPG are inferior in quality to PNG shots then this isn't the fault of JPEG but of Siglent because they set the JPG parameters in their firmware to low quality.

[rf-loop]

d scenery or what ever art pictures using JPG but this is not suitable for this kind of technical document images.

That is not correct. JPEG can provide a similar image quality as PNG, but it depends on the parameters (i.e. quality level, compression factor) that are used for generating them. If the screenshots in JPG are inferior in quality to PNG shots then this isn't the fault of JPEG but of Siglent because they set the JPG parameters in their firmware to low quality.
[/quote]

Not at all if we look result file size.

Original TFT BMP is 1843254 bytes.

Perhaps need some lesson?  PNG is a lossless compression format when JPG is lossy.

You can try your self some exercises if you never have done.

After convert SSA  1.8M BMP to PNG so that PNG file size is around 30k. Look result. (file size depends what you have in picture, because it IS lossless. But also this is why it pack well this kind if images where image do not have nearly anything)
Then do same using conversion to JPG so that result is around same file size. Look result.

But, if you take normal camera image situation is very very different if look PNG and good quality JPG file sizes. In this case JPG win 10-1. Also in normal photograph, it is not so criticalk (usually( if we loose something.. 

And vice versa... in this kind of SSA etc technical images, mostly PNG win JPG like 10-1.

These things is best you self study using example google and read.
But in other hand, everyone has the right to choose a religion, and it must be respected.

Of course  if we use very high quality settings for JPG package result is quite good, also for this kind of images, but why we then use packing if file size is big when we can get still better quality  with lot of smaller file size.

Do you think we want out from BMP just because file size.  Best result is with PNG and difference is not even only marginal.

Here attacehed two images.

Exatly same. Same BMP converted ti PNG and JPG. First I think I try pack JPG to ssame size as PNG size here but result was so terrible that then I select around double file size for JPG paackgage result.

BMP is too big here but even with high zooming I can not find difference between this original BMP and this PNG.
But if you zoom in to JPG it is like pixel porridge. And note, it compressed file size is still over 2x PNG

Do not look what image show, look only image quality. (experiment what you see in image is not what you get from Siglent directly, but 10MHz carrier level is 0dBm)


Load, then zoom both 5x and compare.

[Wuerstchenhund]

d scenery or what ever art pictures using JPG but this is not suitable for this kind of technical document images.

That is not correct. JPEG can provide a similar image quality as PNG, but it depends on the parameters (i.e. quality level, compression factor) that are used for generating them. If the screenshots in JPG are inferior in quality to PNG shots then this isn't the fault of JPEG but of Siglent because they set the JPG parameters in their firmware to low quality.

Not at all if we look result file size.

Original TFT BMP is 1843254 bytes.

Perhaps need some lesson?  PNG is a lossless compression format when JPG is lossy.

Maybe you should educate yourself before trying to lecture others?

https://en.wikipedia.org/wiki/Lossless_JPEG

JPEG can be both lossy and lossless, and lossless JPEG exists for over 20 years (it's often used in medical application, and I believe it's also the format used by the US Librabry Of Congress to archive stuff).

You can try your self some exercises if you never have done.

After convert SSA  1.8M BMP to PNG so that PNG file size is around 30k. Look result. (file size depends what you have in picture, because it IS lossless. But also this is why it pack well this kind if images where image do not have nearly anything)
Then do same using conversion to JPG so that result is around same file size. Look result.

Well, if the file size of the PNG and JPEG are identical then this should have already told you that the JPG is very likely created with a very low quality setting. Simply because lossless JPG is less efficient than the much newer PNG format and a lossless JPEG will be larger than a (lossless) PNG at equal quality.

That is if Siglent even uses lossless JPG, because they might very well just use the lossy variant. Still, even the lossy variant should have no problems to come up with similar quality as a PNG file. But that won't be the case if Siglent uses low quality settings for JPG.

But, if you take normal camera image situation is very very different if look PNG and good quality JPG file sizes. In this case JPG win 10-1. Also in normal photograph, it is not so criticalk (usually( if we loose something.. 

And vice versa... in this kind of SSA etc technical images, mostly PNG win JPG like 10-1.

I hope you don't want to suggest that the crappy 1024x600 (0.6Mpx) screenshot with 256 or less colors displayed (your Siglent SSA-X, like most low-cost low-performance test devices, is highly unlikely to use anything over a 256 color palette, it might well be just 16 colors) and large, homogenous areas is a bigger challenge to a compression format than a multi Mpx photograph with true 24bit resolution and complex content with fine-grnulated, nuanced details?

These things is best you self study using example google and read.
But in other hand, everyone has the right to choose a religion, and it must be respected.

What are you on about? This are technical facts, not religion.

Of course  if we use very high quality settings for JPG package result is quite good, also for this kind of images, but why we then use packing if file size is big when we can get still better quality  with lot of smaller file size.

So you realize the simple reason the JPG is worse is because of the settings Siglent has chosen in their firmware?

Besides, the point of why good test equipment offers more than just one file format is that PNG simply isn't always supported, while JPEG and BMP generally are. So when you have to read the screenshot on a platform/software which for some reason or other doesn't support PNG then PNGs advantages of better compression simply ecome irrelevant.

Do you think we want out from BMP just because file size.  Best result is with PNG and difference is not even only marginal.

We went from BMP (which is usually uncompressed and then pretty much a 1:1 copy of the raw image data, and thus will provide the best image quality) to compressed format because BMP files are huge, and for multi-Mpx images become very difficult to handle. JPG is one of the oldest image formats, it supports lossy and lossless compression, and is widely supported. PNG is a younger format which pretty much exists to circumvent the patents that are in one of JPGs compression formats. And because PNG is younger, it's also more efficient than JPEG, so it shouldn't be surprising when a similar quality JPEG file is larger than a PNG file.

Here attacehed two images.

Exatly same. Same BMP converted ti PNG and JPG.

Nice, but useless. The quality of the JPG pretty much depends on the settings (what compression, what compression ratio, and so on) you have chosen when creating the file (or what the converter you used as default settings), so it's pretty meaningless.

However, the point I was trying to make is that if the JPG screenshots look inferior to PNG screenshots then this isn't the fault of the JPEG file format, it's Siglent who is to blame. JPEG works fine for screenshots if the settings are appropriate, which on most test instruments I've taken JPEG screenshots from are.

[nctnico]

BTW: The BMP format supports RLE compression which is perfect for typical test equipment screendumps.

[tautech]

The standard and only image format for all Siglent products to date has been BMP. That, we imagine will remain however there are moves afoot to offer alternative formats as well. As all images are currently captured as BMP they then need be converted to other formats and then there will be some tradeoffs depending on the computation power available in each product.
We've had some behind the scenes discussion as which formats might be offered and the likely phase in time over the product range. Word I have is that other image file formats will gradually be rolled out in subsequent versions of new FW. We hope for a good selection of formats that offer users the image file type that will suit them best.

[rf-loop]

@Wuerstchenhund

Of course if we talk common photographing and image formats and so on...
But, I was talking in this context where we talk SSA image formats.

Primary is that BMP file size is too big.
We want reduce it and still keep all simple and get also good result and even more, format what is really widely supported.
We can of course use JPG lossless but very easy file size is 200 to over 300k.
Is it nice if we get 300k files (yes it CAN do also smaller but I'm not sure where we hit problems with compatibility in all places.
For this kind of images what are SSA TFT we can very easy do PNG. Get very small file size and keep original image quality.

PNG is not at all best for all things.
But for this purpose PNG  is exellent.

If you believe some lossless "jpg" is better you can believe. But I believe it also leads to many complaints.
I don not talk here  photographing and related to these image formats. JPG may give superior compression for photographic or realistic artwork and for lossless purposes LS and 2000 etc.   but  I talk here SSA3000 TFT screen shot images and what is best solution in practice to get good compression with good quality so that also it is supported as widely as possible. And for this PNG is perhaps even best solution.

Microsoft: "JPEG is not suitable for simpler pictures that contain few colors, broad areas of similar color, or stark differences in brightness."

[Wuerstchenhund]

We've had some behind the scenes discussion as which formats might be offered and the likely phase in time over the product range. Word I have is that other image file formats will gradually be rolled out in subsequent versions of new FW. We hope for a good selection of formats that offer users the image file type that will suit them best.

I don't think there's a need to offer some dozen of more or less obscure image format for a simple screenshot. Offering BMP, PNG AND JPG (maybe with some slamm tweaks to improve quality) is perfectly fine and already more than many other instruments offer.

[tautech]

We've had some behind the scenes discussion as which formats might be offered and the likely phase in time over the product range. Word I have is that other image file formats will gradually be rolled out in subsequent versions of new FW. We hope for a good selection of formats that offer users the image file type that will suit them best.

I don't think there's a need to offer some dozen of more or less obscure image format for a simple screenshot. Offering BMP, PNG AND JPG (maybe with some slamm tweaks to improve quality) is perfectly fine and already more than many other instruments offer.

Exactly what we've asked for. 

[rf-loop]

We've had some behind the scenes discussion as which formats might be offered and the likely phase in time over the product range. Word I have is that other image file formats will gradually be rolled out in subsequent versions of new FW. We hope for a good selection of formats that offer users the image file type that will suit them best.

I don't think there's a need to offer some dozen of more or less obscure image format for a simple screenshot. Offering BMP, PNG AND JPG (maybe with some slamm tweaks to improve quality) is perfectly fine and already more than many other instruments offer.

Exactly what we've asked for. 

Exactly!

[tautech]

When I noticed jpeg screenshot file types added in the 7.07 FW I asked Tech support if they would include png and discussion then focussed on these other formats being offered in other instruments......it's coming I was told. 
"But we should not announce it yet as to provide it across whole the range will take some time".


It may not be possible or wise to include all types on some instruments IMO as the compression and subsequent file upload times may become frustrating, just as it is in some competitors equipment and that's well documented in various posts on the forum. It's something R&D will have to work through to decide which instruments support which formats and what might be acceptable compression and upload delays. Only time will tell what Siglent come up with.

[1design]

I think they should focus R&D efforts more into the debugging and performance sectors, far more important for a SA than the picture format. How often except for really basic report did you even use screen shots at work? Usually the data is imported into a SW and then processed there and replotted using the used SW plotting function. I barely ever used the screen shot function in real life applications.

BR

[Wuerstchenhund]

I think they should focus R&D efforts more into the debugging and performance sectors, far more important for a SA than the picture format. How often except for really basic report did you even use screen shots at work?

Pretty much every day. It's the quickest way to document all kind of stuff.

[Bicurico]

I would like to suggest the EasySpectrum application. It makes grabbing pictures much easier and you don't have the hassle of having to copy things over with a USB stick. I know that this only works if the SA is sitting next to the PC. This way you can grap the pure BMP and save it in whatever format you want! Also, you get to use the "3D waterfall" (not sure how to call it properly).

I have a question on the everyday use of the SSA3021X: when I start to play with Frequencies, Span, Amplitude, Sweep and Trace options, after some time everything gets slow. Even if I set the parameters back to reasonable values, the refresh rate is low. If I press the PRESET button, however, things speed up again. So I tried to memorize those PRESET settings and on the next slow down entered them manually: the SA was still slow. Pressing the PRESET made it run faster again.

Perhaps I am doing something wrong and I did not really measure speeds. Can it be that the PRESET button also resets memories, logs, buffers or whatever?

Regards,
Vitor

[nctnico]

I think they should focus R&D efforts more into the debugging and performance sectors, far more important for a SA than the picture format. How often except for really basic report did you even use screen shots at work?

Well I do that all the time because it is easy. Sure you can use external software but you need a PC, set the software up (and this is often a painful process due to drivers, the right NI VISA driver, etc) and then control the instrument from your PC -if that works at all-. It is just much easier to press a button and when making the report you just drag&drop the images into the document from a USB stick.

[KE5FX]

I think they should focus R&D efforts more into the debugging and performance sectors, far more important for a SA than the picture format. How often except for really basic report did you even use screen shots at work?

Pretty much every day. It's the quickest way to document all kind of stuff.

Yeah, that's crazy talk.  A measurement you didn't record is one you didn't make.

[Electro Fan]

I think normal waterfall, "Spectrum monitor" as it is named in SA GUI is included with Option: AMK-SSA3000X.

Advanced Measurements Kit (AMK-SSA3000X)  – Includes automatic measurements for
– Channel Power
– Adjacent Channel Power Ratio
– Time Domain Power
– Occupied Bandwidth
– Includes Waterfall Display

Here  waterfall image how my SSA3000X  display it. It show level only with color.
(I can not test without this AMK-SSA3000X Option because permanent Option can not remove using SSA GUI)
This looks very different what PC software EasySpectrum image @Bicurico  posted  and it looks 3-d waterfall.



@Bicurico posted nice 3-d waterfall using PC software EasyWave (free)






My SSA3000X (FW 7.05,  with Option: AMK-SSA3000X). Image  from SSA display.

(note also image quality when used JPG or PNG where PNG wins like 10-1.  I hope Siglent add this PNG and make it as default. Peoples can take selfies and scenery or what ever art pictures using JPG but this is not suitable for this kind of technical document images. If yoy zoom some details etc you can see what kind of pixel porridge JPG is coocking in details)

Just to clarify, there are two images posted above; the top looks "3D", the bottom looks like a "2D""waterfall".

Is the ability to make the top 3D image freely included with the SA, or is the top 3D image only possible for an extra charge?  Thanks

[rf-loop]

Upper image is from PC, running free PC software named EasySpectrum. 
This EasySpectrum PC software come free with SA  or  it can also free download from Siglent.
This need SSA is connected to PC and PC software EasySpectrum produce this image. This image is from PC display, not from SSA display at all.  This work with standard SSA without any extra Option. It is EasySpectrum PC software function, not SSA function.

EasySpectrum and other EasyXxxx softwares for other Siglent equipments can download here.
http://siglentamerica.com/support_software_14
Before install it, first need install  National Instruments VISA (Ni-Visa 15.0.1)
Also it is free. It can get from National Instruments.

Of course if have good programming skills there can produce what ever images using data from SSA and command SSA using SCPI commands.  These can find in SSA3000X  Programming quide





Lower image is directly from stand alone SSA3000X display (and screen dump to USB stick).   It is one function included in Option AMK-SSA3000X.  This Option is not free and for activate it permanently need buy license key. When buy new SSA there is short trial time for all Options. But these need try out soon after get SSA because all options timers run always when SSA is running, not only if you use option. ( The trial period implementation is not good. My opinion is that counter need count only when Option is used.)

Advanced Measurements Kit (AMK-SSA3000X)  – Includes automatic measurements for
– Channel Power
– Adjacent Channel Power Ratio
– Time Domain Power
– Occupied Bandwidth
– Includes Waterfall Display

[rf-loop]

I have a question on the everyday use of the SSA3021X: when I start to play with Frequencies, Span, Amplitude, Sweep and Trace options, after some time everything gets slow. Even if I set the parameters back to reasonable values, the refresh rate is low. If I press the PRESET button, however, things speed up again.

I have never meet this exept in situations when I have changed some parameter what affect speed. I mean example if I have adjusted VBW, after adjusting it is manually forced until I change its mode back to auto. Now If I forget this and if I do some other things and then just example change span. Then I wonder why this is so slow. Until I realize that oops... VBW is narrow. There is many this kind of things what do not return back to auto where SSA set itself better value when user select some other parameter. Also then, example sweep mode, in some (most) cases sweep is slower that FFT. If you normally use in auto sweep mode and then you select manually sweep mode sweep. Then  change freq, span etc it may work much slower than usually if in sweep mode auto where is select FFT or (default) speed optimize instead of accuracy.

Of course I have not tested all possible combinations (human life is not enough for all possible combinations), so it is still possible there is some situation where it live some process on what slows system.  When find this, it is good to document exacxtly these kind of findings so that same can repeat example by other user or in Siglent.

One I know. If use AMK option "waterfall". When it start from empty waterfall image it is more fast. After it collect more and more data it goes  slower, until there is full waterfall image, after then it run with this speed.

Also if turn on Demod and set demod time long then forget demod on and go to look some other things. But demod time still make "pause" between sweeps. (in this case there is AM or FM displayed in left side)
If there is example 500ms demod time and then you go to some other things and your sweep time is example 10s. You do not perhaps note this extra pause. But then you adjust BW and or span etc and you expect much more fast sweep. Then you wornder why 24ms sweep is so slow. It keep demod time pauses (and you are not using it but forget it on)


Needs a lot of experience before the device is so familiar with that note of all the factors that affect each other and how they interact with each other.
Know your device - it is always important.
Then, when you know or seriously suspect that something is really wrong, it is good to document it well. So well that the problem may be repeated anywhere at any time by other user and I'm sure Siglent want also check all documented problems carefully.

[rf-loop]

About LIMIT function. One small example. Of course it can use for many things.
Also it have both, upper limit and lower limit.
It can use in two kind of limits. Just simple level limit line over span or user defined shape limit line.
Limit trace can fully edit and one limit line may have up to 100 points what each have level and frequency. Limit line is based data points. When limit is used and If you change ref level, scale and span etc limit line follow its own data points even if they are out from screen. Data points keep values what you have made.

In this example perhaps most simple case where is user edited limit line (in this case only 7 data points) These limits can store and use later of course. (file type .lim)

Some times it is nice to leave SSA watching some frequency band if there exists some times some signal. Of course for this can use Max-Hold function but it is not always optimal.  There was continuous signal and highly variable level so I made limit line what give more room for this signal.
Then I set limit rule so that SSA stop when limit test fail.
Now it run as in first image and after some signal hit limit, it stop like in next image. Also Marker was set for trace highest peak.

Limit line editing is easy and fast after short exercise. This siglent control knob is really good, accurate and fast also . Not at all like some oscilloscopes encoders. Also finger feel these small steps. Finger point to the small hole, and then a rotation, it goes like HP in the old days. (but I do not believe it is optical with as good bearings like old times HP)


Running, no any signal hit limit.



Signal hit limit, stop. (marker is not because hit limit, it is because at this time this signal peak is highest)

[nctnico]

That is not correct. JPEG can provide a similar image quality as PNG, but it depends on the parameters (i.e. quality level, compression factor) that are used for generating them. If the screenshots in JPG are inferior in quality to PNG shots then this isn't the fault of JPEG but of Siglent because they set the JPG parameters in their firmware to low quality.

Not at all if we look result file size.

Original TFT BMP is 1843254 bytes.

Perhaps need some lesson?  PNG is a lossless compression format when JPG is lossy.

I did some tests and with the JPG quality set over 90% you won't see the differences between BMP and JPG but still get a more than 10 times reduction in size. What is most worrying is that nobody at Siglent gave any thought about the JPG compression setting or which file format to support. It is all trial & error with the end user in the loop.

[Bicurico]

I have a question on the everyday use of the SSA3021X: when I start to play with Frequencies, Span, Amplitude, Sweep and Trace options, after some time everything gets slow. Even if I set the parameters back to reasonable values, the refresh rate is low. If I press the PRESET button, however, things speed up again.

I have never meet this exept in situations when I have changed some parameter what affect speed. I mean example if I have adjusted VBW, after adjusting it is manually forced until I change its mode back to auto. Now If I forget this and if I do some other things and then just example change span. Then I wonder why this is so slow. Until I realize that oops... VBW is narrow. There is many this kind of things what do not return back to auto where SSA set itself better value when user select some other parameter. Also then, example sweep mode, in some (most) cases sweep is slower that FFT. If you normally use in auto sweep mode and then you select manually sweep mode sweep. Then  change freq, span etc it may work much slower than usually if in sweep mode auto where is select FFT or (default) speed optimize instead of accuracy.

Of course I have not tested all possible combinations (human life is not enough for all possible combinations), so it is still possible there is some situation where it live some process on what slows system.  When find this, it is good to document exacxtly these kind of findings so that same can repeat example by other user or in Siglent.

One I know. If use AMK option "waterfall". When it start from empty waterfall image it is more fast. After it collect more and more data it goes  slower, until there is full waterfall image, after then it run with this speed.

Also if turn on Demod and set demod time long then forget demod on and go to look some other things. But demod time still make "pause" between sweeps. (in this case there is AM or FM displayed in left side)
If there is example 500ms demod time and then you go to some other things and your sweep time is example 10s. You do not perhaps note this extra pause. But then you adjust BW and or span etc and you expect much more fast sweep. Then you wornder why 24ms sweep is so slow. It keep demod time pauses (and you are not using it but forget it on)


Needs a lot of experience before the device is so familiar with that note of all the factors that affect each other and how they interact with each other.
Know your device - it is always important.
Then, when you know or seriously suspect that something is really wrong, it is good to document it well. So well that the problem may be repeated anywhere at any time by other user and I'm sure Siglent want also check all documented problems carefully.

I think I have figured out great part of the SSA3021X, but naturally there is still much to learn.

Perhaps I just forgot to set one parameter and thus got the feeling that the refresh rate was slow.

I will try to pay attention to this and if it really gets slow, I will try to find out what is the cause, so that it can be reproduced. Right now I am not sure if it is my mistake.

All in all, I do like the device very much. It is a great step up in regards to the best of my field meters with spectrum function.

I find the resolution amazing, showing (often the right word is revealing) signals in a detail I couldn't see before with my existing devices.

Another aspect, that I haven't read in this thread and might be of interest to some potential buyers: the screen is huge! When I first got the SSA3021X I was surprised about the size of the screen. Much bigger than what I would have guessed from the pictures. And the resolution of 1024x600 really gives enough room to properly display everything.

While I understand that professionals who have to comply with tight specifications need to use a premium spectrum analyzer (with a price tag that can easily have an extra digit), I would say that this SSA3021X will do the job for most hobby or casual professional application (if specs like bandwidth are within requirements) hands down. At the current retail price, this is a real bargain.

Regards,
Vitor

[Performa01]

Isn’t it funny how the picture format for screenshot files has become a major topic here in a thread about a spectrum analyzer – quite obviously just because it looks like a good opportunity to bash Siglent once again.
But Siglent needs not give a thought about JPEG compression settings, since this is an SA and not a camera of any kind, hence it outputs technical graphs. Consequently, the screenshots come as BMP and not JPG – most folks who have followed this thread and actually read what has been posted would have been aware of that by now.

Of course, we want to keep our documents small, so we use some tool on our PCs to convert the bitmaps to some compressed file format. RF-Loop has pointed out that PNG is the best file format for this, since it’s lossless and the resulting files are tiny nevertheless. He tried to get a similar file size with JPG, but failed, as the result was plain unusable. Even with twice the file size, the BMP was still rather ugly. Hardly a surprise – as PNG is able to boil the 1126kB BMP from a typical scope screen down to just 16kB – that’s a 70:1 ratio, without any loss of image fidelity.

The lack of PNG file format might be a little inconvenience – for those who don’t do any post processing anyway, e.g. optimizing the gamma for the document or adding some remarks – but still a very insignificant issue for sure.

But yes, it would be nice to have PNG as alternative file format for screenshots in the future, but that’s it then.

I don’t have an SSA3000X, but even so I would much prefer the firmware development team to concentrate e.g. on the TG control, so that it can accurately measure steep filters in SNA mode, instead of tinkering with picture file formats.

[nctnico]

Isn’t it funny how the picture format for screenshot files has become a major topic here in a thread about a spectrum analyzer – quite obviously just because it looks like a good opportunity to bash Siglent once again.
But Siglent needs not give a thought about JPEG compression settings, since this is an SA and not a camera of any kind, hence it outputs technical graphs.

Siglent has a long history of getting the format for saving screenshots totally wrong. So yes, this is an issue! Every manufacturer of test equipment has been using JPEG, PNG or even compressed BMP for decades. So it is just dumb they choose to use JPEG and not care to have a look whether the output is 'grainy' or visually perfect. People do need screendumps for reports and copying large BMPs into a document isn't going to help to keep the size down so the document can be e-mailed.

[radar_macgyver]

Isn’t it funny how the picture format for screenshot files has become a major topic here in a thread about a spectrum analyzer – quite obviously just because it looks like a good opportunity to bash Siglent once again.

As an added data point, Anritsu's spectrum analyzers (with a 10 dB higher price tag) also use JPEG with compression turned up when saving screenshots. This was seen on a Spectrum Master 20 GHz hand-held, with the latest firmware.

Some folks just have an axe to grind. 

[tautech]

Isn’t it funny how the picture format for screenshot files has become a major topic here in a thread about a spectrum analyzer – quite obviously just because it looks like a good opportunity to bash Siglent once again.
But Siglent needs not give a thought about JPEG compression settings, since this is an SA and not a camera of any kind, hence it outputs technical graphs.

Siglent has a long history of getting the format for saving screenshots totally wrong. So yes, this is an issue!

Since you don't even own one and have often declared your dislike for Siglent products after a previous unfortunate experience we can only assume you are trolling again.