Hisilicon Has Very Advanced Soc ... But The Surveillance Industry Insists On Using 4 Year Old Chips

Have you looked at the kinds of things Hisilicon has?

http://www.hisilicon.com/en/Media-Center/News/Hi3231V52X_DPT

I'd like to buy a chip and replace the board/SoC on a unit and play with modern equipment. This is what monopolies produce; intentional or apathetic stagnation.

Night imaging? Sony. Cheap imaging? HiSilicon. Is there any reason to expect them to provide us with modernity's tech? See what chipsets these units are using.

Maybe this is an unfair and ignorant appraisal of a market I know so little about, I don't know how little I know. But dayum, why can't we have flip chips which upgrade? :-)

Reactions:

John Honovich

IPVM

What is so interesting about that chip? I ask that genuinely since the press release does not say much nor do I see anything that really jumps out as such a breakthrough.

More importantly, the chip was just press released in July. Is it even shipping to manufacturers yet? Even if it is (and often these things have delays), it will take the manufacturers some time to incorporate / test / optimize, release into new products, etc. Moreover, my understanding is that typically chip production ramps up slowly and only to large customers first (since they get priority).

I'd like to buy a chip and replace the board/SoC on a unit and play with modern equipment.

You can't do that. Also, why do you think the current SoCs used in cameras is so antiqued and not modern? I would assume this one is better than the SoCs from 18 months but not so revolutionary that it would drastically change results.

I know this is blunt but I am also trying to be realistic about what to expect.

Related, where are you with your UHD / 4K cameras that you asked about 2 months ago? Are those working well or?

Reactions:

Truman HW

In reply to John Honovich

The nM size is related to impedance, which correlates to heat generally.

I'd venture that 90% of latency is the SoC, and not the network unless it's incredibly congested and isn't dedicated to networking.

This chip does UHD at SIXTY FRAMES per second... not 30 at the expense of smart features... but sixty in H.265. That's awesome. And if it's able to do so while retaining the smart features, it's nothing short of a bad mamajama in my book.

I'd also add, that these things cost far less to manufacture than sold for - but, when you include marketing, support, unsold inventory, price wars, I get it. As many have said, it does become an ugly race to the bottom.

Trickling out the product that's best has never really served an industry unless there was, at minimum, silent colluding to maintain pricing.

A MacBook Pro that sold for $1200, cost $268 to manufacture. Yes, their stores are beautiful, but a commensurate PC is a FRACTION of the price. Why? DIRECT competition.

This is the reason however, that we're all members here at IPVM.

The 'parity' of experts telling what they really think.
The reviews! which settle fact from fiction.
The incredible resource of answers to problems.

It's just too bad it's not less expensive for manufacture's goods to be subjected to an apples-to-apples comparison in the way IPVM doed. Maybe manufactures should be required to submit to standardized tests. IP-SAT. :)

Reactions:

Undisclosed #1

IPVMU Certified

I'd like to buy a chip and replace the board/SoC on a unit and play with modern equipment.

Good luck with that!

The chip you are referencing would appear to be a SOC for a television/display.

Reactions:

(2)

John Honovich

IPVM

appear to be a SOC for a television/display.

Lol, well that would be another reason :) Good eye.

But let's assume it was a SoC for an IP camera, all the comments I made above still apply.

Reactions:

Truman HW

I happen to own a business in which we do BGA rework, which is a sub-feature of SMT repair; so replacing an IC package that has matching modular outs and package size isn't unreasonable to re-solder. :)

Reactions:

Undisclosed #1

IPVMU Certified

...replacing an IC package that has matching modular outs and package size isn't unreasonable to re-solder.

Truman, regardless of how you connect it, that chip is the exact opposite of what you want.

You need to read pixels off an imager and compress and encode and stream.

Not read a stream and decode, decompress and write to a display.

Maybe run reverse polarity ;)

Reactions:

Truman HW

I do plan on making some videos of my tech repair things... maybe replace the NAND on an iPhone or iPad - and after you see the micro soldering via microscope, etc, we can all judge.

Reactions:

Truman HW

Hi "Undisclosed" :-) ....

So that chip isn't what receives the data uncompressed - with the job of compressing it?

[What] chip then, is charged with that task on the surveillance cameras, and what the f is the hisilicon chip on surveillance cameras charged with?

Is there a homunculus in there pre-watching this stream, which requires said chip you allege is decompressing the stream? ?

Reactions:

Truman HW

Camera

https://goo.gl/KbrHbg

Lens + DSP ??

https://goo.gl/C5uGmi

Why do these look like the Hisilicon device is somehow related to the recording of the device?

Also, these are always called.... CODECS. Meaning, COder, DECoder, no? Meaning, definitively, does both ?

Kinda like the way your MODEM modulates, AND DEMODULATES.. ?

Reactions:

Undisclosed #1

IPVMU Certified

Kinda like the way your MODEM modulates, AND DEMODULATES.. ?

A modem is a bi-directional communication device.

The day when your VMS starts sending h.264 streams *to* your camera is the day when your camera will need a decoder.

Reactions:

Truman HW

Okay, undisclosed, what then, is your hypothesis of what the Hisilicon chip's role is, in the cameras that it so ubiquitously resides?

Reactions:

Undisclosed #1

IPVMU Certified

Sure, but only after you tell me what camera the Hi3231V52x 4K Panel Driver chip ubiquitously resides in

Reactions:

John Honovich

IPVM

#1 is objecting to the specific part you list in the opening post.

Hisilicon, of course, makes SoCs for IP cameras, just not the part you list. For example, here is their upcoming 8K/30fps chip they promoted at CPSE 2017. See our CPSE 2017 report for more comments on Hisilicon.

By the way, why no comments or questions from you about Ambarella? Ambarella is typically ahead of Hisilicon when it comes to IP camera SoC development.

Reactions:

(1)

Truman HW

This was the point I was sloppily getting at. I apologize for the erroneous shill. :)

Reactions:

Truman HW

•Nov 13, 2017

HI3519V101 is a better example....

Reactions:

Undisclosed #1

•Nov 13, 2017

IPVMU Certified

It is better, right on!

But if the premise of the discussion is:

Hisilicon Has Very Advanced Soc ... But The Surveillance Industry Insists On Using 4 Year Old Chips

then this chip doesn’t really make the strongest case, since it appears to be featured in several models currently available, for instance:

Reactions:

(1)

Truman HW

Yes, but no.

Where are the major manufactures using this? That's the frustration; that if any of the big guys DID use this, they'd charge a fortune for it. But that's germane, as there's obviously no chance in that... because some of them don't even think they need to get down with H.265.

The "yes" however, is that I'm VERY glad to see a few vendors I know who have good deals on the newer HiSilicon chip... though, I don't know if there're any new(er) Sony sensors.

Reactions:

Undisclosed #1

IPVMU Certified

...though, I don't know if there're any new(er) Sony sensors.

Truman, these chips aren’t sensors...

Reactions:

Truman HW

DUDE (since you hide under anonymity). Are you ... serious? You're really explaining that a DSP and an image sensor are different? :) Thank you.

Anonymous, without researching it,

1. Why is it that smaller MP counts are used for low light?
2. Why is the cost of a larger sensor not linearly priced to manufacture?
3. What was the big 'loss' of digital images vs. analogue?
4. How many variables regulate exposure?
5. What kind of image is viewed on your display when you open a JPG?
6. If you view a 1MB JPG, how many MB of VRAM (shared or not) is used?
7. Whats the difference between NAND as an SSD vs. RAM ?

(I do realize these are remedial)
For the fun of it, let's go over some acronyms/abbreviations:

EEPROM
IDE
NNEXP
SCSI
iSCSI

Why is VoIP both better and worse than standard telephony?
Contrast a NAS and a SAN

I PROMISE, I won't need to look up info to test you.
I'll know if you cheated.
Please, see how quickly you can answer these... And - to your question and my tangent of realizing the answer to my own question... the assumption I hold.

For a second there, I thought HiSilicon and their DSP which works as the codec was the same exact thing as the sensor. In fact, I thought the sensor just, encoded itself. I mean, doesn't it?

Maybe I worded something that lead you to believe that I have spectacularly remedial hardware knowledge, somehow.

But, what I was driving at is, most of the sensors I see from companies that publish their components have been the IMX 178 or 226... and that hasn't changed in about 2 years - even though sony does have some sensors that are higher res, low(er) light, etc. But much like hard drives were a bottleneck for decades in the computing world, I can see how sensors have to wait for DSPs which can keep up with the heat generated by encoding to something that's 30% the size of a H.264 file... and remuxing/transcoding whatever the raw image is 24/7, is a lot of work from devices which aren't afforded the "room" to make noise in terms of a fan... or in which it'd be tolerated for them to fail every 18 months due to the heat the chip endures, quietly.

So DSPs do have their work cut out for them in catching up and exceeding the buzzwords that capture the financiers interest in surveillance; 4k.

Personally, I'd rather modular camera systems, where you can upgrade SoC/DSP... the sensor, the local storage (M.2 or SSD over SD)...

And in this way, you could cool the DSP, which would allow it a faster clock, which would afford it the capacity to manage more FPS. Or to prioritize silence, and to merely replace the module which failed -- which is so obviously a conflict of interest for the big companies.

Reactions:

Undisclosed #1

IPVMU Certified

Off the top of my head...

1. Why is it that smaller MP counts are used for low light?

Lower fill factor/shot noise of smaller wells reduce photon capture efficiency
2. Why is the cost of a larger sensor not linearly priced to manufacture?

yields decrease with bigger wafers

3. What was the big 'loss' of digital images vs. analogue?

Not sure what you want, maybe compression? Though not a necessary component of digital images per se.
4. How many variables regulate exposure?

Aperture, shutter speed, gain (or film speed for old school cameras)
5. What kind of image is viewed on your display when you open a JPG?

kinda vague don’t you think, Maybe a bitmap?
6. If you view a 1MB JPG, how many MB of VRAM (shared or not) is used?

Varies maybe 10x as much or more

7. Whats the difference between NAND as an SSD vs. RAM ?

Vague again, Non-volatile vs volatile is maybe what you want?

Reactions:

(1)

Truman HW

Not bad.

1. Why is it that smaller MP counts are used for low light?

I don't know if your answer is false.

I believe that, if the sensor size is static, the fewer divisions, the more sensor area remaining to receive photons.

We disagree (on item 1.) for at minimum, the order of the cause. The effect however of increasing the noise is a fact. Perhaps your statement is intended to have the same effect as mine; it's just more convoluted and jargonistic. I assert that, all things being equal, the simpler answer wins.

2. Why is the cost of a larger sensor not linearly priced to manufacture?

You've answered only the effect; you've left the cause untouched. If the increase of price is ubiquitous, then it's based on physics, and not psychology. WHY do the yields decrease? Is it merely because dividing 1000/50 is a smaller number than dividing it by 10? Who would know from your answer?

The issue is that there is an error rate in the substrate per area; and as that area increases exponentially, increasing from say, a DX to an FX sensor means that only 1/3 as opposed to 10/13 are "usable." While, obviously, companies will use algorithms to determine how to maximize profits with respect to demand per wafer.

If errors increase exponentially, cost/yield is obliged to decrease proportionate to the errors per wafer.

3. What was the big 'loss' of digital images vs. analogue?

How would one compress analogue - without making it digital?

Latitude.

That is, the range in which you could adjust the brightness or darkness retrospectively to get properly exposed film. While the latitude of film is large, that of digital is small. Medium format digital cameras do pretty well... Nikon's like the D800 are pretty good at getting proper exposure from seemingly dark images... but Film (last I checked) was simply superior.

4. How many variables regulate exposure?
Three things. (You're correct.) Aperture, Shutter speed, Sensitivity - however it's measured.

5. What kind of image is viewed on your display when you open a JPG?
Correct.

6. If you view a 1MB JPG, how many MB of VRAM (shared or not) is used?
Principle is correct - number is arbitrary. (nicely done).

7. Whats the difference between NAND as an SSD vs. RAM ?

Pretty much; The distinction (probably what you meant by volatility) is whether the data is lost when the charge is lost.

Do, you think that I think a sensor is a signal processor, or that it codifies the image received?

Reactions:

Undisclosed #1

IPVMU Certified

1. Why is it that smaller MP counts are used for low light?

Lower fill factor/shot noise of smaller wells reduce photon capture efficiency

Perhaps your statement is intended to have the same effect as mine; it's just more convoluted and jargonistic. I assert that, all things being equal, the simpler answer wins.

Fill factor (from my answer) is the simple, industry standard term of art for what you are trying to describe in your:

I believe that, if the sensor size is static, the fewer divisions, the more sensor area remaining to receive photons.

as Wiki says

The fill factor of an image sensor array is the ratio of a pixel's light sensitive area to its total area.

2. Why is the cost of a larger sensor not linearly priced to manufacture?

yields decrease with bigger wafers

You've answered only the effect; you've left the cause untouched.

I see nothing asking for “cause” nor “effect”, only “why”, therefore my answer “yields decrease with bigger wafers” is fully explanatory. Suggest you rewrite the question.

btw, yield is defined as the ratio of acceptable/total parts, therby encompassing your error rate in substrate per area. Nor do I see any further cause defined by yourself, such as expounding upon the limitations of the Czochralski process in mono-crystalline fabrication.

3. What was the big 'loss' of digital images vs. analogue?

Not sure what you want, maybe compression? Though not a necessary component of digital images per se.

That is, the range in which you could adjust the brightness or darkness retrospectively to get properly exposed film.

Word to the wise; on this site analog (without further qualification) refers to NTSC Analog cctv cameras. If you mean film photography, you need make that clear. As to your “how would you compress Analog without making it digital?” Using an Analog compressor of course. Compression takes an analog signal and reduces the analog, bandwidth required thru dynamic range reduction or frame rate reduction, as seen in slow-scan tv, or in POTS where frequencies are stripped and dynamics are limited.

7. Whats the difference between NAND as an SSD vs. RAM ?

Non-volatile vs volatile is maybe what you want?

Pretty much; The distinction (probably what you meant by volatility)...

Again, non-volatile as in NVRAM is the stand industry term, not my invention, but one you should consider adopting.

Do, you think that I think a sensor is a signal processor, or that it codifies the image received?

Honestly Truman these questions are questions that a general PC tech might ask. There are plenty of people who know the answers to these questions that would not know anything about the imaging pipeline at the chip level.

But I think you do know the difference; your post-hoc reply gratuitously incorporating Sony part numbers indicates a depth of knowledge seldom seen around here.

Keep up the good work!

Reactions:

Truman HW

Thank you for the kind words. As far as the industry terms, I see your point, and I will evaluate it.

I have this belief that, "If you can explain it to a non-tech person, you're pretty competent. If you can explain it to a child, you're a master."

But yes, amongst peers, the bandwidth improves when you use the vernacular.

Wafer-efficiency; the number of areas that are 4x the original size (double in each dimension) with say, an error margin of 5% is going to go from ~1/20 to 1/2. Of course, you're going to be able to maximize the yield from all the areas suitable for the smaller sensor, which mitigates the cost -- but also forces a secondary reason to raise the price of the larger sensor... the more large sensors you make, the more small ones you make. Sell em off? Create your own competition. (Ad infinitum).

AND, if you're the same undisclosed from earlier, your contrast of a SAN and a NAS was so good, I might copy it for future explanations. I have little experience with the, but I know some of the fundamentals.

What's TRULY exciting... are the new DSPs:

Hi3559AES Hi3559AV100
2xA73@2GHz + 2xA53@1.1GHz, H.265 Main 10 / H.264, 8K@30fps or 4K@120fps, 2/4/6 channel image stitching
HDR10/14bit ISP pipeline, Sensor input up to 36MP, PCI-o 2.0 + USB 3.0, High-end smart vision

DONT EVEN ASK ME... I assume that the

Hi3559AV100 is the encoder for the camera feed:

Hi3559AES seems like its meant to be the receiver, with (captain obvious) AES encryption.

But, check that out, it does 4k at 120FPS. Without.... a fan??

What do you do with the USB here? lol

I'm a tech, retail business owner, learns about something like I'm going in to that business... and enjoy giving people logic board repairs for half of what logic charges people. :)

I would however be happy to hear your take on the naming convention, and what else we can learn from it.

Reactions:

Undisclosed #1

IPVMU Certified

EEPROM Electric(?) Erasable Programable Read only memory
IDE integrated drive electronics
NNEXP don’t know
SCSI small computer systems interface
iSCSI Internet scsi

Reactions:

Truman HW

Close - (Principle: Wrong/Recall of acronym: Good) 'Electrically': Conjugation changes meaning.
Good enough - Trick question, two answers.
Correct - This was my cheat test. :) Falsifying a fake abbrev. is border impossible
Good.

Reactions:

Undisclosed #1

IPVMU Certified

Try mine: How Well Do You Know Your Acronyms?

Reactions:

Undisclosed #1

IPVMU Certified

Why is VoIP both better and worse than standard telephony?

How about VOIP better because it can be cheaper and worse because it adds latency.

Contrast a NAS and a SAN

NAS are drives on a LAN often shared by multiple computers, SANs are drives usually connected to a single host via fiber. NAS drives use higher level file abstractions, SAN uses lower level methods. With a SAN the host “knows” where on the device sectors reside, with a NAS it knows only the metadata for the file.

Reactions:

Truman HW

Latency vs. Bandwidth; the economics of voice, if we're assuming within the US - are free vs. free. Unlimited. VoIP is 128KB often, and I believe standard voice is 28k, which is an odd number (not an exponential result of "2")... and yes, latency [and] continuity are superior with voice. Van Neumann is who I believe extrapolated the notion of what constituted 'data' or a signal, which was billable while working for AT&T. He's a polymath worth reading about if you ever have the time. We all benefit today from his paper on information theory.

SAN vs. NAS: More than sufficient to demonstrate knowledge of the systems.

CS major?

Reactions:

Trisha (Chris' wife) Dearing

IPVMU Certified

DUDE (since you hide under anonymity).

I am not hiding, I find it easier to stay focused on what is being said rather than who is saying it when one or both parties are undisclosed. But since you object, I am disclosed for this post.

Are you ... serious? You're really explaining that a DSP and an image sensor are different? :) Thank you.

No offense Truman, but for the better part of this discussion you didn’t realize that the chip you were touting in the OP was not even a chip meant for cctv cameras. Even though I plainly declared it right away.

So, yes, it seemed to me a likely conclusion that you were conflating an SOC with a sensor, in your following statement:

The "yes" however, is that I'm VERY glad to see a few vendors I know who have good deals on the newer HiSilicon chip... though, I don't know if there're any new(er) Sony sensors.

Your later explanation of sensors are waiting on SOCs is specious at best, and in no way could be inferred from only your initial statement above.

Reactions:

Truman HW

In reply to Trisha (Chris' wife) Dearing

Actually, I DID acknowledge... that you nit-picked that it was the decoder - not the encoder. I even explained where the name CODEC came from (in the hopes of mitigating your committed belief that I'm an unintelligible ignoramus for copying the wrong part number from a page that discussed BOTH). I didn't known the naming convention from which to defend against so pivotal an error. I'm SURE you didn't have to look up what that part number corresponded to, you just knew it - right?

Would you do me the honor of providing a succinct, clear explanation of the naming convention HiSilicon uses? I can't BELIEVE how stupid I am. I mean - the fact that I didn't memorize their naming convention, by which, I could defend against such pedestrian mistakes as copying and pasting the wrong term. And I completely see how you'd be disabled from discussing the point, which I think someone with an IQ of 108 could infer:

And then!? Then?? I REALLY did it. Shhhh. My... opinion....!?? evolved? over time!

I get your point; no one should ever evaluate or articulate their thoughts more clearly. First draft. Always. It's what all authors and software companies do. Right? All those stupid revisions! Just ruins the product.

Where is my DOGMATISM!? Huh?? Howma gonna git ta heaven without more dogmatism!??

Are you, one of my fellow Americans who thinks "flip flopping" is a bad sign in a candidate when it's not for political gain, but as a byproduct of accruing information and modifying their beliefs to be in better harmony with the stupid data? Evolution's just a damned theory!

Just respond relentlessly to the title or a mistaken part number. DON'T update your responses based on new info, and for Christ's sake! The earth is 6,000 years old! How many times have I got to say it?? It'd be "specious" to say otherwise - and that's a damned commandment. GET WITH IT Chris.

My Guess: Hard Drives cause consumers more dissatisfaction than CPUs.

For the last decade - and potentially since the advent of the PC, hard drives were/are, the primary source of perceived 'slowness', frustration, and allowed manufactures to mislead you to believe it was every part but the one it was -- because no affordable technology existed to solve that pain.

This could be analogous to the race between MP and compression/hard drive space... or the ability to review it smoothly. Consumers will respond to 'consumerisms' like 4k - and manufactures will entice them with it. But, from the continuity of compatibility for their devices to the amount of heat generated by compressing data to a size that's practical to store and decode is still more challenging than consumers are warned about.

If we can exclude the servers people's phones/computers use, an '06 PC or Mac with a fast SSD will be more pleasurable throughout most uses, even in a Core Duo, than a much newer computer with an i7-7700 but a crappy hard drive that's in pre-failure. And, if you look at the ratio of HDD performance vs. CPU performance over time; that is, the number of GB/s a CPU can process, vs. the number Mb/s the cable supported, let alone, the health of the miniaturized record player - I'd say it's been this way for decades, I just don't have the data before the last decade to personally assert. Worse, consumers, thinking their WHOLE computer had "gotten old and slow" -- bought new computers, when only a hard drive was needed.

Take the 1.1GHz MacBook Retina, from 2015. FINE. Why? The SSD went around 800MB/s, and had a very high IOPS.

It's possible similarities exist in other domains.

Since you want youtube style comments, this is what you get:

Were you asserting that DSPs don't exist in cameras? That all cameras and all manufacturers impeccably offer the newest one? If your goal is to discuss ideas... rather than try to belittle other members, then why not skip the condescension and move straight to your beliefs on the topic?

Please, DO tell me... WHAT were you genuinely confused by? WHAT did you think my point actually was? I... often understand people when they speak a language I don't... by picking out just a few words that are cognates. Assuming you don't have Asperger's, were you REALLY disabled from plausible understandings? Or were you humping my leg because I used a decoder instead of an encoding part number...?

Augmenting your beliefs is a tenuous interpretation of what specious means. Again, you think I was arguing for a modified belief of my original statement, as though that's a duty to uphold. You have a very bizarre set of ideological beliefs to insist on MY meaning, rather than allowing it to mean what I'm RIGHT HERE to explain should you feel it's ambiguous.

If I pulled a bunch of ICs off a motherboard we fix (we do SMT repair after liquid or drop damage) - and you didn't know what all of the part numbers meant, would that suffice as your IQ? Aren't principles more interesting to discuss than spelling errors. The one, non-linguistic reference caused the entirety of your beliefs about me to be prejudiced?

I elaborate and tell you more about my thoughts - and you think I'm trying to imbue the "tagline" of the discussion with meaning that I'd concede if you'd merely ask, wasn't even on my mind when I wrote it. DID IT HAVE TO BE? I thought the "reply" button was for that purpose. Why isn't the "Post" button renamed "Concrete" ... ?

Did you know you were going to say specious when you originally charged me with the a misnomer infraction? HA! SEE!? You're lying! You DIDN'T know you'd say specious later, ergo, your earlier statement is (who knows what the hell to fill in here because I have NO idea what principle that illustrates!).

Does Axis, Pelco and other similarly positioned and prominent companies currently utilize in any of their product - DSPs that are released within the last 18 months - as well as listing said DSP publicly in their specifications?

My point was and is -- that the major manufactures that people here would regard as top dogs...

1. aren't focused on getting the newest DSPs
2. if they do, don't acknowledge it in their specifications
3. Maybe they do what apple does and make their own?

Either way, they play by different rules.

Chris, aka, Anonymous:
Please, tell me IF the lack of a part number ruined the above's capacity to be clear.
Did what I say within the italics relate sufficiently to the title?

Reactions:

Undisclosed #1

IPVMU Certified

And I completely see how you'd be disabled from discussing the point, which I think someone with an IQ of 108 could infer:

Yes, your correct, I found it hard to discuss your point, due to my intelligence. This is the way I saw it, translated to a subject more accessible, like baseball. Links are to the original posts.

HW: Why are batters using 4 year old batting technology when there are better and newer bats out there, like the Spaulding #ABC123?

Me: Ok, but that ABC123 is a baseball glove, so...

HW: I own a company that does some sports equipment work and I can tell you that we can replace pretty much any equipment for any other.

Me: Yes, but still that equipment is the opposite of what you want. You need something to hit the ball with not catch a ball.

HW: Yeah, well what do you think a lacrosse stick does? It catches and throws the ball.

Me: If the time comes that the batter needs to catch the pitch instead of hit it, then it would make sense.

JH: I think that the objection is to your suggestion of a specific part which happens to be a glove, not a bat.

HW: Ok, #ABC789 is a better example.

Me: Yes, that a new bat, but wasn’t the premise of the discussion why don’t we use the newer bats? That bat is used in several available models.

HW: Well, I meant major manufacturers...

Reactions:

(1)

(2)

Truman HW

I totally enjoy your specious analogy. :p It IS amusing.

But AGAIN. I harken back to the point you've already received; MODEL NUMBERS AREN'T VERNACULAR. And I would bet money that I could easily find some tech, some company, who's naming convention who'd as of yet MEMORIZED. And thus, KNOWING that you were talking about a glove instead of a bat, wouldn't be immediately obvious.

And, someone operating in good faith may have said, "Do you mean the HI3519V101? Because the one you're talking about is a TV decoder..." To which, I'd have replied, "Ah, yeah, excuse me, I grabbed the wrong model item." And the discussion of the PRINCIPLE would have moved on.

Again, I had no idea you had Asperger's, and that literalism vs. inference was the threshold of your capacity to engage in a conversation.

Again (part 2): Can you please, without cheating tell me the naming convention for HiSilicon products? And really - I'd rather do this in Skype, so that in real time I can quiz you to demonstrate that you don't know ALL companies product-naming-conventions. And absent such knowledge, one wouldn't intuitively realize they'd erred in identifying a product.

I realize the flowchart of humping one's leg to point out that, "neener neener neener, you said the wrong model!" is an extremely enticing option to 8 year olds who's ego depends on putting others down to elevate their own standing, relativistically. But another line coming off that box in the flowchart, is inferring what someone means, and discussing the IDEA, and not the specific part -- as I'd be willing to bet that "most modern technology NOT used" doesn't end there, with DSPs. Or, would you like to "short" that "stock's" idea?

Yes, you can defend the minor points within your choices... but macroscopy will show that you also had a choice of a high road, but you behaved immaturely. And YES, I admit it - I've followed you down to the youtube-comments-level, all too willingly. But I'll also freely admit, I think you're far too smart for this. You don't need to correct someone's [spelling] (read: arbitrary memorization) to be seen as a "smart guy."

Just saying, "Hey Truman - I think you mean the HI3519V101... but, to that end, there are some smaller companies that have started using them, even if not the Elite companies public endorsement of the products... if at all."

Then, you'd have been knowledgable, followed the principle rather than the decoy (no, it wasn't intentional... before you pretend I'm alleging it was) and would have gracefully accomplished everything with concision, maturely.

Do you really think you went about this in a way that wouldn't have made more than 20% of people defensive and offended?

PS:
Spelling, as well as the arbitrary memorization of the trivial, is the virtue of the dumb.

Smart people discuss ideas.

Reactions:

Undisclosed #1

IPVMU Certified

And, someone operating in good faith may have said, "Do you mean the HI3519V101? Because the one you're talking about is a TV decoder..."

Truman, in my first message to you I said:

The chip you are referencing would appear to be a SOC for a television/display.

which, as you mention, would have been a prime time for you to say:

Ah, yeah, excuse me, I grabbed the wrong model item.

But, you didn’t, did you?

Not for a week.

Why didn’t you say something when I said it was a TV chip?

Really. Why?

Reactions:

Undisclosed #1

IPVMU Certified

...in the hopes of mitigating your committed belief that I'm an unintelligible ignoramus for copying the wrong part number from a page that discussed BOTH).

What page exactly would that be?

Reactions:

Truman HW

I was using my other laptop as mine was fixed. :( Sorry. I'll try to upload the image.

I was reading through pages on Unifore - and either they were both on the same page, or both in my recent history... and I was overly quick about flipping back. Obviously, it's my fault.

Reactions:

Undisclosed #1

IPVMU Certified

Thing is that searching for that part returns three results, the Hisilicon product page, the graphic I linked to, and of course your discussion.

So, how exactly did you get that link?

Something like “Latest 4K new technology HiSilicon chip” works.

Reactions:

Truman HW

So you missed the part where I said that I went back in my history and grabbed what I thought it was. This was a week ago. I don't remember what I had for food.

And... acknowledging your nit-pickery wasn't high on my list; I worked 70 hours this week... that was.

I'm a business owner. Sometimes we get busy. I apologize for not getting on here earlier and pacifying your ego's entitlement to being right this week. I realize what a huge impact that had on your stats, and I won't do it again. My deepest apologies.

Will you be okay?

Reactions:

Undisclosed #1

IPVMU Certified

I apologize for not getting on here earlier and pacifying your ego's entitlement to being right this week.

Its okay; mentally I had already “chalked it up” a while back.

Will you be okay?

Yes, unless you ace my acronym test. Then I’ll really have a lot of soul-searching to do :)

Reactions:

Undisclosed #1

IPVMU Certified

So you missed the part where I said that I went back in my history and grabbed what I thought it was.

Actually, you first said

“the hopes of mitigating your committed belief that I'm an unintelligible ignoramus for copying the wrong part number from a page that discussed BOTH).”

But there is no page that has BOTH. When I pointed that out you then said you went back and grabbed it from your history.

Even if that is the answer, the question is how did you get to the page at all? Since there are only 3 ways to get there?

Were you browsing for TV SOCs on HiSilicon’s site? Maybe without knowing what you were looking at?

Further evidence is provided by your statement that you were reading about the 101 chip on Unifore’s site, which makes the point:

In this article, we briefly introduce you the Hi3519 + Panasonic MN34120 4K network camera solution, because we expect this solution will be deployed in plenty consumer 4K cameras and sports cameras, as well as security surveillance cameras soon.

And that’s just with the Pana imager.

So, really, what point did you want to discuss, after reading that article? The opposite point?

Reactions:

Truman HW

So, Asperger's and OCD? Which psych meds do you take?

I own quite a few computers, and as I said, I was either reading through pages on new chips, or a page which discussed both.

FYI, you may want to use wildcards in your search, because sometimes they may not include the conjugation of the 100/101, as those are variants of the main generation I think.

Speaking of unanswered questions:

Weren't you going to explain the NAMING CONVENTION HiSilicon uses, which would be the means by which someone could avoid referencing the wrong chip?

Earnestly, I just trusted that the point would be the same irrespective the WIDGET I referenced, which was explicitly about cameras and their respective DSPs

I think I also said that I don't know a lot about this industry... but that something akin to replaceable chips would be awesome.

If you'd like to interrogate me more, perhaps you can get a subpoena and a warrant? Barring those two, I'm going to ignore you. Fascinating as you'd be to someone studying abnormal psychology, you bore me.

Reactions:

Undisclosed #1

IPVMU Certified

I own quite a few computers, and as I said, I was either reading through pages on new chips, or a page which discussed both.

A link is worth a thousand insults.

Reactions:

Undisclosed #2