Doug_Park

And to be clear. I think you have been professional.

Don't stop complaining. We like people who are willing to work with us to help improve the program. Some of our most effective testers are those that complain the most. The only thing I ask is that you try to do so in a professional manner.

It is probably the PDF viewer doing anti aliasing of the fonts and lines. Normally you want this when you send it to a printer because at that resolution the result will look smoother. At lower resolutions some people perceive the result as blurry rather than smoother. We don't do this in Chief when we display on screen because it is a lot slower to draw.

The dual graphics card option is actually pretty good as it allows you to draw less power when running on a battery. I have that option in my Dell M3800. Both the Intel and NVIDIA graphics work well but the NVIDIA is faster, although not as much faster as you might expect.

I used to want bigger screens on my laptops. But after lugging a heavy 17" laptop around for several years I opted for lighter smaller screens. My MacBook Air has an 11" screen and is very portable. I even took it with me on my cross country bicycle trip last fall. I also have a Dell m3800 with a 15" 3200 x 1800 screen which weighs a little less than twice as much as the Air. I don't miss the large screen and if I do really need more space I can easily plug in a large secondary monitor. But your needs are different than mine so you will need to decide for yourself what you want.

Dell has lots of good options as does HP. A lot of laptops today have both an NVIDIA and Intel graphics. This is a good option as it allows you to either go with the faster high power usage NVIDIA or the slower Intel that uses less power depending on your power needs. So either go with one that has just an NVIDIA or one that has both and you should be OK. Quad core is best. Dual cores are OK, but will not do ray tracing nearly as fast or other operations. I would highly recommend going with a Quad core. Displays can be a bit tricky. Some of the high end laptops are using 4K (high DPI) displays. Chief supports these well, but some other applications don't. You will pay more for one of these displays. The benefit is very crisp fine lines. The down side is some applications may not display things correctly. 8GB is probably a good minimum target memory today. 16GB is where I would likely want to be to ensure a 5-6 year lifespan of the laptop. I would highly recommend getting an SSD disk. These are very expensive in 1TB sizes. But you can get a external USB drive if you wish. Windows 8.1 is an excellent choice from a reliability standpoint. It also boots a lot faster.

Go to the NVIDIA control panel. It has a place to set when to use the two different cards. Both are actually pretty fair performers for Chief so if you are running on battery power and want to conserve power using the Intel graphics is not a bad choice. On my laptop I have Chief set to use the NVIDIA card. but let other apps use the less power hungry Intel graphics.

Unfortunately, we don't support rotating elevation camera views. It would be nice to do.

There are a number of high end laptops with touch screens that will work. Tablets like the surface pro are good options but they are not going to be nearly as fast as a high end laptop. They are likely faster than a 3 or 4 year old mid range laptop though.

Yes, the arc creation modes are something that are probably underutilized when drawing arcs. Getting familiar with them should pay off if you draw a lot of curved stuff. They work for walls too.

The icons for those indicators are the same as the command they are associated with. There are a handful that we turn on to help people know they are in a mode that is possibly not one they want to stay in permanently. I can't remember all of them. Rebuild terrain, edit object parts, concentric mode, and some of the other somewhat obscure behavior modes are some.

So the 2670 has 12 cores. and the 2697 has 14. The turbo speed is also different. It is hard to judge what would be best without doing some benchmarking. But 24 cores vs 14 should give you more throughput for ray tracing assuming the memory can keep up. Either option would be the envy of most users of Chief though.

Modeless interfaces are important. I don't want to overstate how much gain you would get, which is less than what most would think. One must consider everything when estimating how much you would save. Consider how one would provide 3D previews of objects as we do now. The cost of keeping a 3D view of an object up to date all the time would add up. Also consider the cost of populating a dialog. The expense of the time required to open a dialog currently would be paid every time one selects a new object even if you didn't need it. While this is small, usually less than 1 second, the time adds up and could, if not done very well, lead to an overall additional sluggishness of the program. If one is to provide the sometimes dozens of input fields that our objects have then you are still going to pay the price of a click to switch panels or to scroll to the location where the setting you want to change is located. For something very simple like a line one could put everything on the same panel. We already provide some modeless interfaces. Such as tool buttons that effect changes to an object, like the center tool and the ability to enter dimensions. To be clear, I am in favor of adding modeless interfaces. Look for more to be added into the program as we move forward.

Some of the low end versions of Windows don't support multiple processors. The pro versions do. Mac also supports multiple processors. Chief will use as many processors as the OS exposes.

Sorry if I'm being dense here. But I don't understand what you mean. Are you saying that a modeless interface is not a good idea or something else?

Yes. We detect all the CPU cores in the machine regardless of the number of processors. Hyper threading can help some, but it depends a lot on the operation being performed whether it is a significant gain. It is never a 2X gain.

Improvements to the UI have been very significant in X6. I'm not surprised that some of the improvements have not been noticed because we worked hard to make the transition as painless as possible while at the same time improving how things work. There is still a lot to do though. The modeless interface is a good and important thing. It would produce a small but measurable improvement in productivity in some cases. However, there would be at least a short term decrease in productivity while the new interface is learned. And in other cases there would be no advantage while at the same time making it easy to accidentally change a property of an object. For example, someone complained in another thread how it was too easy on this forum to accidentally change the font size. Still it is something we should do and will pursue in an order that hopefully won't cause a great deal of transitional grief.

This isn't a problem unique to CAD. In word processing programs there are styles that can be used on a per paragraph basis, selection basis, etc. By leveraging these you can write a document and then with a few changes of the styles change the look of the entire document. I personally leverage this when I'm writing a long formal document. However, this requires knowing how to use these tools. Typical users of word processors don't want to be bothered by this so will select the text they typed and change how it looks using the easy to access tools. Given these two ways of working in a word processor the first is overall much faster in the long run, but requires learning something that is not obvious, and is also hard for some to understand. Both ways of working have value. The value of each varies depending on how one wants to work. Our view has always been that setting up layers and layer sets to be the primary control of how things look is most efficient in the long run so we have worked hard to make this method more efficient. It seems like some effort to make the more simple tools easier to use and access is important.

It seems like there are two ways of working being described here. 1) Let the layer control line style, color etc. In this way of working one would create a layer for the specific color, style, and weight. Then it is a simple matter of just changing the layer and start drawing. To control display a layer set is created to turn on/off the collection of layers that need to be displayed. 2) Put everything on 1 layer so that turning a layer on/off controls the collection. In this model one needs to change the color, style, and weight of each line. In this style it makes more sense to want to change things up front as is requested. It is less flexible if you ever want to change how some things look. Neither method is wrong. Chief is designed to make the first method easier. I may be misinterpreting how you are working, if so please let me know.

I agree that a modeless interface is better. It isn't particularly easy to retrofit in Chief, although it is easier to do now than it was before we released X6. The importance of this is a question to ask. Is it more important than fixing stairs or live updates to layouts or making framing work better? It is a question we have to answer and so far the answer seems to be that other things are more important. But I could be wrong.

Lot's of good input on this thread. Many of the employees here read ChiefTalk. This forum is one of many sources of information about what we need to put into the program. Our support and sales teams talk to customers a lot which provides another good source of information. Our engineers communicate with customers from time to time as well to get input on certain things. Our marketing team also does surveys as well as competitive analysis. We take all of this information and discuss it internally to come up with priorities.

Lot's of ways of doing this. I did this quickly by setting up one window, setting the multiple copy distance to get what I wanted and then used multiple copy to build the windows. Then group select the windows and center them in the room. You can also block them into a mulled unit. At the spacing they are at they will automatically mull together. But if you block them you can then copy paste them to new locations if you would like.

For me having a touch screen isn't an advantage. I have on on my latest laptop but pretty much never use it. Possibly because I switch computers frequently and don't have that option on all of them. But that might be different for others.

Probably, but would cost development time and might involve purchasing a third party library to provide the ability to record to more formats. I certainly wouldn't want to implement the recording ourselves.

The formats we save to are currently limited by availability on the OS. Mac provides some methods that Windows doesn't.