I wish we could turn off the automatic measurement feature of the oscilloscope, for teaching purposes. 

 This probably sounds weird, but I think it important that students get familiar with the waveforms and actually count divisions.  This gives a feel for how many digits are really significant, that noise can be part of signals, that it helps to increase the sensitivity to fill as much of the screen as possible  and maybe notice artifacts.  e.g. Students might be asked to measure frequency response of a circuit with a uA 741 op amp.  If the signal level chosen is too high and students don't actually watch the waveform they may not realize the numbers come from the slew-rate limit and not the frequency response.  I actually recognized the problem of the seductiveness of digital displays in another lab where students used a lock-in amplifier with both a digital display and a d'Arsonval meter.  The voltage went through large changes in value, and students were happily writing down the numbers when the instrument was on a totally inappropriate range.  So for the first electrical-engineering lab I think forcing the students to rely on measuring the waveform themselves and not relying on the scope to do it for them gets them started well.  

In the signal integrity business at these data rates, we are always looking for higher bandwidth and a lower noise floor. Because this product is an 8-bit real-time scope, I would like a lower noise floor and have it go to a 9-bit or 10-bit resolution. That would be beneficial.

On the source side, their AWGs don't have enough bandwidth. They need to increase the bandwidth of their AWGs, meaning they have to have a faster sample rate. Though, they generally have plenty of resolution. A 100 gigasample per second AWG would be excellent.

Tektronix sells a lot of options. Those options are a bit of confusing at times. It would be nice if the package included everything so I wouldn't have to figure out what I need to buy.

In terms of room for improvement, it's a powerful scope. But, the power of the computer inside is weak. Sometimes I send a lot of data and I'll feel the lag. The computer is slow. Eventually it provides the answer. I would like to see them improve the processing power.

It needs one button and every time you push it, you get whatever you need. Label it W.I.M.(what I mean) and advertise it as a wimp interface. Naturally the GPIB has only one command and does whatever you mean it to do.

I would like to see a lower price.

I know you can get the Bit Masks but it's generally really expensive to get the eye diagrams and Bit Masks with it. I wish they were included or cheaper.

There was a learning curve, which was a little tricky at times. Getting used to the GUI took time because we had to learn the different features, but once we learnt it, it was good.

Their hardware piece for their test engagement boxes builds on its own. The chips from some of the ICs technically build it themselves. Then, they look for a market for ICs from high end analog to augment their solutions. It is important for Tektronix to own the digitizer chips in the ICs and roadmap.

Tektronix should continue to invest in their own hardware, since most companies cannot do what they do due to size and power. They should continue to invest on their own process going forward.

I am not utilizing the full bandwidth of this product. However, Tektronix is organizing the buttons, menus, etc., so it's not cluttered and user-friendly. 

We are looking for an Intel platform with the newest chipset.

They have to support the new industry standard.

There are other devices out there which are more capable, offering more features, with as good quality tech support.

If they could make the user interface even more intuitive that would help. If they could make it easier to read, to know where the buttons are, that's all I need.

I would like if they could have more customer support engagement.

Going forward, they should keep up with the industry's speed for oscilloscopes.

We would like more features for the CAN bus decoding, such as importing DDC files, which allows us to take the raw data and convert it into engineering units. So, if I wanted to look at speed from a vehicle instead of looking at a hexadecimal or binary values on the oscilloscope it can load the DDC file and display the actual speed in miles per hour.

Its connectivity and ability to use it with PCs and smart phones needs improvement.

They need to add more memory to this machine because at this time the machine can support random simulation to the power of 15 but realistically, in requests from our customers, we need it to the power of 31. It's typically memory which is the issue.

There is room for improvement when it comes to the screen size. I don't really have a lot of complaints but screen size would be the biggest thing.

I use microservices, but you cannot really probe it, since the technology is not really there.

I would like to see an app from this company for test fixtures. It's very difficult for us to find test fixtures, so it would be helpful if they were to develop them.

I would like to see more integration into machine learning because that's what we try to apply it to.

The ruggedness could be improved.

I want the system to have a wider bandwidth because high-speed continues to move faster, so there is more bandwidth to cover. 

I would like to see some WiFi capabilities. I would like to upgrade equipment, and if I was in the U.S., I could do that with WiFi capabilities.